Title of Invention	RELAY STATION AND METHOD FOR ENABLING RELIABLE DIGITAL COMMUNICATIONS BETWEEN TWO NODES IN A WIRELES RELAY BASED NETWORK
Abstract	A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) and a method (700) are described herein that enables reliable digital communications to occur between two nodes in a wireless relay based network (600, 800, 900, 1000 and 1100). The wireless relay based network includes a first node (602, 802, 902, 1002 and 1102) that transmits information in coded/modulated digital communications to a second node (604, 804, 904, 1004 and 1104) via one or more relay stations. And, each relay station is capable of: (1) receiving (702) a coded/modulated digital communication from the first node; (2) computing (704) a plurality of reliability values for a plurality of information symbols or coded symbols in the received coded/modulated digital communication; and (3) transmitting (706) a coded/modulated digital communication that has the computed reliability values embedded therein to the second node.

Title of Invention

RELAY STATION AND METHOD FOR ENABLING RELIABLE DIGITAL COMMUNICATIONS BETWEEN TWO NODES IN A WIRELES RELAY BASED NETWORK

Abstract

A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) and a method (700) are described herein that enables reliable digital communications to occur between two nodes in a wireless relay based network (600, 800, 900, 1000 and 1100). The wireless relay based network includes a first node (602, 802, 902, 1002 and 1102) that transmits information in coded/modulated digital communications to a second node (604, 804, 904, 1004 and 1104) via one or more relay stations. And, each relay station is capable of: (1) receiving (702) a coded/modulated digital communication from the first node; (2) computing (704) a plurality of reliability values for a plurality of information symbols or coded symbols in the received coded/modulated digital communication; and (3) transmitting (706) a coded/modulated digital communication that has the computed reliability values embedded therein to the second node.

Full Text	BACKGROUND OF THE INVENTION Field of the Invention The present invention relates in general to the telecommunications field and, in particular, to a relay station and method for enabling reliable digital communications to occur between two nodes in a wireless relay based network. Description of Related Art Manufacturers and operators of wireless relay based networks are constantly trying to develop new ways to improve the reliability of digital communications transmitted between two nodes. The traditional wireless relay based networks described below each include one station (node A) that transmits information in coded and modulated digital communications to another station (node B) via one or more relay stations (RSs). The relay station can be part of a base station (BS), a mobile station (MS) or a stand-alone relay station. The nodes A and B can be a BS, a MS and/or a relay station. And, the digital communications could be either uplink communications (link from MS to BS), downlink communications (link from BS to MS), MS to MS communications or BS to BS communications. Examples of different traditional wireless relay based networks and some of their respective drawbacks are briefly discussed below with respect to FIGURES 1-5. Referring to FIGURE 1 (PRIOR ART), there is shown a block diagram of a traditional wireless relay based network 100. The wireless relay based network 100 includes a first node 102 (node A) that transmits information in coded and modulated digital communications to a second node 104 (node B) via one relay station 106. As shown, the first node 102 transmits a channel coded and modulated signal So(t) over a first link 108 that has a complex channel coefficient h0 such that So(t) is received at relay station 106 as R1(t)=h0S0(t)+n1(t), where ni(t) is a noise sequence. The relay station 106 is a repeater that generate a signal S1(t) which is an amplified version of the received sequence and is shown as S1(t)=AR1(t). The relay station 106 then transmits the generated signal Si(t) over a second link 110 that has a complex channel coefficient h1 such that S1(t) is received at the second node 104 as R2(t)=h1S1(t)+n2(t), where n2(t) is a noise sequence. There are several drawbacks associated with this type of wireless relay based network 100. First, the amplification of R1(t) at relay station 106 amplifies not only the signal R1(t) but also the noise n1(t) caused by the first link 108. Secondly, the relay station 106 is not very efficient when it amplifies R1(t) since there is a large amount of redundancy in So(t) due to channel coding. Referring to FIGURE 2 (PRIOR ART), there is shown a block diagram of another traditional wireless relay based network 200. The wireless relay based network 200 includes a first node 202 (node A) that transmits information in coded and modulated digital communications to a second node 204 (node B) via one relay station 206. As shown, the first node 202 transmits a channel coded and modulated signal So(t) over a first link 208 that has a complex channel coefficient ho such that So(t) is received at relay station 206 as R1(t)=h0So(t)+n1(t), where ni(t) is a noise sequence. The relay station 206 then decodes, re- encodes, re-modulates and transmits a signal S1(t)=AlSoest(t) over a second link 210 that has a complex channel coefficient h1 which is received at the second node 204. This is a good solution whenever the relay station 206 makes a correct decision when it estimates and re- encodes So(t). However, when the relay station 206 makes an incorrect decision, the second link 210 further propagates the information errors and increases the bit error rate of the signal Si(t) transmitted to second node 204. Referring to FIGURE 3 (PRIOR ART), there is shown a block diagram of yet another traditional wireless relay based network 300. The wireless relay based network 300 includes a first node 302 (node A) that transmits information in coded and modulated digital communications to a second node 304 (node B) via one relay station 306. As shown, the first node 302 transmits a channel coded and modulated signal So(t) over a first link 308 that has a complex channel coefficient ho such that So(t) is received at relay station 306 as R1(t)=hoSo(t)+n1(t), where n1(t) is a noise sequence. The relay station 306 then decodes, re- encodes, re-modulates, amplifies and transmits a signal S1(t) over a second link 310 that has a complex channel coefficient h1 which is received at the second node 304. In this example, the relay station 306 checks the correctness of the decoding of So(t) using for example a cyclic redundancy check (CRC) and only re-generates, re-modulates and transmits S1(t)=AlSo(t) in case of correctness, otherwise the relay station 306 simply amplifies and retransmits the received signal as S1(t)=A2R1(t). This solution is problematic since the relay station 306 needs to make a hard decision on the information symbols in S1(t) instead of having the second node 304 make that hard decision. It is well known that in a communication chain it is advantageous not to make hard decisions until late as possible in the chain. As described below in detail there are some known ways to avoid making hard decisions in relay stations which include using a second relay path or using an automatic repeat request (ARQ) protocol. Referring to FIGURE 4 (PRIOR ART), there is shown a block diagram of yet another traditional wireless relay based network 400. The wireless relay based network 400 includes a first node 402 (node A) that transmits information in coded and modulated digital communications to a second node 404 (node B) via one relay station 406. As shown, the first node 402 transmits a channel coded and modulated signal So(t) over a first link 408 that has a complex channel coefficient ho to relay station 406. The relay station 406 then transmits S1(t)=Q(R1(t)) which is a quantized base band representation of R1(t) over a second link 410 that has a complex channel coefficient h1 to the second node 404. In this solution, the second link 410 generally has a larger bandwidth and uses a different air interface protocol than the first link 408. This solution is problematic in that the quantized base band signal S1(t) contains a large amount of data about phase and noise that does not necessarily need to be sent to the second node 404. Referring to FIGURE 5 (PRIOR ART), there is shown a block diagram of a traditional wireless relay based network 500. The wireless relay based network 500 includes a first node 502 (node A) that transmits information in coded and modulated digital communications to a second node 504 (node B) via two relay stations 506a and 506b. As shown, the first node 502 transmits a channel coded and modulated signal So(t) over a link 508a that has a complex channel coefficient ho to relay station 506a. At the same time, the first node 502 also transmits So(t) over a link 508b that has a complex channel coefficient ho to relay station 506b. The relay stations 506a and 506b know or can estimate their respective channel coefficients ho, ho, h1 and h1 on links 508a, 508b, 510a and 510b. As such, relay station 506a can transmit S1(t)=Aconj(h0)conj(h1)R1(t) over link 510a to the second node 504. And, relay station 506b can transmit S1(t)=Aconj(h0)conj(h1)R1(t) over link 510b to the second node 504. As is well known the signals R1(t) and R1(t) because of the scaling and phase shifts involved can be added coherently in a maximum ratio combining sense when they are received at the second node 504. This scheme can be extended to three or more relay paths. Even with multiple relay paths there is still a reliability problem since each relay station 506a and 506b may make a different decision in decoding signals R1(t) and R1(t) which makes it difficult for the second node 504 to coherently combine signals S1(t) and S1(t). One possible solution when there are multiple relay paths is for each relay station 506a and 506b to check an error detecting outer code in signals R1(t) and R1(t). If either R1(t) and/or R1(t) checks out OK then the respective relay station 506a and/or 506b transmits the corresponding S1(t) and/or S1(t), otherwise the respective relay station 506a and/or 506b does not send anything or they can send an amplified version of the respective received signal(s). The problem with this solution is that it relies on the following assumptions that (1) at least one relay station 506a or 506b has correctly decoded signal R1(t) and R1(t) and (2) that the corresponding second link 510a or 510b for such relay station 506a or 506b is sufficiently good to enable the decoding of signals S1(t) or S1(t). Accordingly, there is a need for a new signal processing solution that can be implemented in a relay station which addresses the aforementioned shortcomings and other shortcomings of the traditional wireless relay based networks 100, 200, 300, 400 and 500. This need and other needs are satisfied by the wireless relay based network and relay station of the present invention. BR1EF DESCR1PTION OF THE INVENTION The present invention includes a relay station and a method that enables reliable digital communications to occur between two nodes in a wireless relay based network. The wireless relay based network includes a first node that transmits information in coded/modulated digital communications to a second node via one or more relay stations. And, each relay station is capable of: (1) receiving a coded/modulated digital communication from the first node; (2) computing a plurality of reliability values for a plurality of information symbols or coded symbols in the received coded/modulated digital communication; and (3) transmitting a coded/modulated digital communication that has the computed reliability values embedded therein to the second node. BR1EF DESCR1PTION OF THE DRAWINGS A more complete understanding of the present invention may be had by reference to the following detailed descR1ption when taken in conjunction with the accompanying drawings wherein: FIGURES 1-5 (PR1OR ART) are block diagrams of five different traditional wireless relay based networks; FIGURE 6 is a block diagram of a wireless relay based network and a relay station in accordance with one embodiment of the present invention; FIGURE 7A is a flowchart illustrating the steps of a preferred method for using the relay station shown in FIGURE 6 in accordance with the present invention; FIGURE 7B is a diagram illustrating how a Log-APP filter can implement an APP algoR1thm which is one way to help enable the method shown in FIGURE 7A; FIGURE 8 is a block diagram of a wireless relay based network in accordance with another embodiment of the present invention; FIGURE 9 is a block diagram of a wireless relay based network in accordance with yet another embodiment of the present invention; FIGURE 10 is a block diagram of a wireless relay based network in accordance with still yet another embodiment of the present invention; and FIGURE 11 is a block diagram of a wireless relay based network in accordance with yet another embodiment of the present invention. DETAILED DESCR1PTION OF THE DRAWINGS ReferR1ng to FIGURES 6 and 7A, there are disclosed preferred embodiments of a wireless relay based network 600 and method 700 in accordance with the present invention. The wireless relay based network 600 includes a first node 602 (node A) that transmits information in coded and modulated digital communications to a second node 604 (node B) via a relay station 606. Like the traditional wireless relay based networks 100, 200, 300, 400 and 500 shown in FIGURES 1-5, the relay station 606 can be part of a BS, MS or a stand- alone relay station. The first node 602 and the second node 604 can be a BS, a MS and/or a relay station. And, the digital communications could be either uplink communications (link from MS to BS), downlink communications (link from BS to MS), MS to MS communications or BS to BS communications. However, the relay station 606 of the present invention has a functionality shown in method 700 that is a marked improvement over the functionalities of the traditional relay stations 106, 206, 306, 406 and 506 descR1bed above with respect to the traditional wireless relay based networks 100, 200, 300, 400 and 500. The relay station 606 basically enables information to be reliably transmitted between the first node 602 and the second node 604 by receiving (step 702) a coded and modulated digital communication 607a which was transmitted from the first node 602 over a wireless link 608 that has a complex channel coefficient h0 The relay station 606 then computes (step 704) a plurality of reliability values for a plurality of symbols in the received digital communication 607a (R1(t)). There are many ways of which two are descR1bed in detail below as to how the relay station 606 can compute the reliability values for the symbols in the received digital communication 607a. In the first way, the relay station 606 utilizes a maximum a posteoR1 (MAP) filter 612a to compute the reliability values for code symbols based on a code structure of the received coded digital communication 607a. For more details about the MAP filter 612a reference is made to the following articles all of which are incorporated by reference herein: • J. Lodge et al. "Separable MAP "filters" for the Decoding of Product and Concatenated Codes" Proceedings of IEEE International Conference on Communication, pp 1740-1745, 1993.* • I. Land et al. "On the Interpretation of the APP AlgoR1thm as an LLR Filter" ISIT 2000, Sorrento, Italy, June 25-30, 2000.* * It should be noted that in the literature the acronym MAP (maximum a posteR1oR1) can be called APP (a posteR1oR1 probability, Log-APP or Max-log-APP. FIGURE 7B is a diagram showing how a Log-APP filter 612a can implement an APP algoR1thm that helps perform steps 702, 704 and 706 in the method 700. It should be appreciated that the notation used herein is the same notation as used in Land et al. where the log-likelihood values for the code symbols are denoted by L(X) and the log-likelihood values for the information symbols are denoted by L(U) with superscR1pts + or - depending if the corresponding reliabilities are considered before (-) or after (+) the Log-APP filter 612a. The log-likelihood value L(X) for a binary code symbol X is In (P(X=+1 \| received data) / P(X=-1 I received data))(see Land et al.) As such, it follows that equally probable symbols then have log-likelihood value (L-value) of zero and a highly likely '+1' have a large positive L-value, and conversely a highly likely '-1' have a large negative L-value. Most often the information symbols are equally likely from the receivers perspective before processing, hence L(U")=0 is often assumed which further simplifies the calculations in the Log-APP filter 612a. After processing the updated L(X+) and/or L(U+) these values can be embedded into transmitted radio signal (see step 706 in FIGURE 7A). In the most basic and preferred form either information or code symbol reliabilities are embedded, but of course they can be combined as well. If only code symbol reliabilities are to be used, of course the reliability calculator (Log-APP in the example) need not calculate information reliabilities or vice versa. FIGURE 7B shows the case where all information is used in the most general form. The Log-APP filter 612a can be interpreted as using the code structure to 'amplify' the reliability and the more amplification there is the more reliable the symbols are up to a limit decided by the code parameters. If only L(X") and L(X+) are used then the Log-APP filter 612a can be seen as a very non-linear log likelihood filter which outputs values in the same domain with the same type of values and code symbols as the output but with an improved signal to noise ratio. It should be appreciated that since the input and outputs are of the same form (L(U) and/or L(X)) then in some cases, after a first application of the APP algoR1thm that the output L(X+) can be connected to the input L(X-) and the APP algoR1thm can be run again with these new values as input. The same follows for L(U)-values. An additional functionality and advantage of using the Log-APP filter 612a shown in FIGURE 7B is that it makes use of the redundancy and structure of the channel code in the received digital communication 607a to clean-up or filter that digital communication 607a and redistR1bute noise to unreliable parts in the digital communication 607b (S1(t)) transmitted to the second node 604. It should be appreciated that the Log-App filter 612a does not necessaR1ly regenerate a valid code word but computes a code symbol by code symbol reliability value. In the second way, the relay station 606 utilizes a soft output channel decoder 612b to compute the reliability values for information symbols based on a code structure of the received coded digital communication 607a. To accomplish this, the soft output channel decoder 612b can employ for example: (1) a maximum a posteR1oR1 (MAP) algorithm; (2) a soft output Viterbi algoR1thm (SOVA); (3) a Log-MAP algoR1thm; or (4) a Max-LOG-MAP algoR1thm. For more details about the soft output channel decoders 612b reference is made to the following articles all of which are incorporated by reference herein: • P. Robertson et al. "Optimal and Sub-Optimal Maximum a Posteriori Algorithms Suitable for Turbo Decoding" European Transaction on Telecommunications (ETT), Vol. 8, No. 2, pp. 1-16, 1997. • J. Hagenauer et al. "A Viterbi Algorithm with Soft-Decision Outputs and its Applications" IEEE Globecom, pp. 1680-1686, 1989. The reliability values that are computed for the code symbols or information symbols could have anyone of a wide variety of formats. For example, a binary '1' symbol could have a reliability value +1.0 if there is a 100% certainty of a T. And, a binary '0' symbol could have a reliability value -1.0 if there is a 100% certainty of a binary '0'. Moreover, the reliability value can be 0.0 if the bit is completely unreliable (T and '0' equally probable). In another example, a reliability value of+0.2 could indicate that a binary T symbol is more likely to be correct but a +0.8 would indicate an even stronger conviction of a binary '1' symbol. It should be appreciated that there could be more possible reliability values than symbols in order to be able to exploit 'soft' information. For example, a reliability value of {+1.0, 0, -1.0} could denote three possible reliabilities for two symbols '0' and T. In another example, a reliability value of {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} could denote seven possible reliability values for '0' and '1' symbols and also provide a better resolution. The seven possible reliability values could be represented by a 3-bit code. After the relay station 606 computes reliability values for either the code symbols or information symbols within the received digital communication 607a, the relay station 606 then transmits (step 706) a coded and modulated digital communication 607b that has the computed reliability values embedded therein over a wireless link 614 which has a complex channel coefficient h1 to the second node 604. There are a vaR1ety of ways for embedding the reliability values in the transmitted digital communication 607b (S1(t)). For example, the reliability values can be explicitly embedded into the transmitted coded digital communication 607b by actually transmitting data representing those reliability values. In another example, the reliability values can be implicitly embedded into the transmitted digital communication 607b by using the reliability values to modulate the amplitude (and/or phase) of the transmitted digital communication 607b. Several different examples are descR1bed below on how the reliability values can be embedded into the transmitted digital communication 607b. Example #1: The reliability values can actually be embedded in the transmitted digital communication 607b if one increases the bandwidth or bitrate of the transmitted digital communication 607b. And, instead of transmitting reliability values of'0' or '1' which takes a single bit one could get a better resolution by tripling the bit rate and transmitting three bits denoting one of seven reliability values {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} for each of the symbols in the digital communication 607b. In this example, the relay station 606 would act as a supportive processing element for the second node 604 (e.g., user terminal). This option is especially attractive if the wireless relay based network has a very high capacity second link 610 like the ones shown in FIGURE 9. Example #2: If multiple reliability values {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} for each of the symbols are embedded in the transmitted digital communication 607b, then something has to be increased like the bandwidth as descR1bed in example #1. Another option is to increase the size of the signal constellation from 2 to 8 (for example) to embed multiple reliability values associated with each symbol in the transmitted digital communication 607b. For example, a binary phase shift key (BPSK) may be used when two reliability values for each symbol are embedded in the transmitted digital communication 607b. If there are seven reliability values as described in example #1 then 8-PSK is an option. Example #3: To implicitly transmit reliability values one can use different signal constellation sizes (modulation schemes) for symbols with different reliabilities such that more energy per symbol is spent on the reliable symbols in the transmitted digital communication 607b. Example #4: One could generate the transmitted digital communication 607b and modulate it using the reliability values for the corresponding bits in the digital communication 607b. For example, one can amplitude modulate the transmitted digital communication 607b with the reliability of symbol at time t, REL(t) and transmit the coded digital communication 607b (e.g., S2(t)=REL(t)*Si(t)). In this case, the reliable parts of the digital communication 607b are transmitted with higher power than the low reliability parts of the digital communication 607b. This is done because there is no reason to waste energy on already unreliable bits since the reliability of a bit can only decrease in the transmission chain. Example #5: The reliable parts of the digital communication 607b could use a longer signal time occupation signal than the one for unreliable parts. This is another way of ensuring that reliable parts are transmitted with higher total energy. An advantage of this example when compared to example #4 is that the amplitude of the transmitted digital communication 607b is now constant which is sometimes desirable. It should be appreciated that any modulation scheme (amplitude modulation, phase modulation) implies an expansion of the bandwidth on the link 610 to the second node 604. Again, it should be appreciated that there are other ways that are known or can be subsequently developed which can be used to embed reliability values into the transmitted digital communication 607b in accordance with the present invention. Referring to FIGURE 8, there is shown a block diagram of another embodiment of a wireless relay based network 800 in accordance with the present invention. The wireless relay based network 800 includes a first node 802 that transmits information in coded and modulated digital communications to a second node 804 via two relay stations 806a and 806b. As shown, the first node 802 transmits a channel coded and modulated signal 807a (So(t)) over link 808a that has a complex channel coefficient ho to relay station 806a. At the same time, the first node 802 also transmits digital communication 808a' (So(t)) over a link 808a' that has a complex channel coefficient ho to relay station 806b. The relay stations 806a and 806b respectively perform the signal processing associated with method 700 and transmit digital communications 807b and 807b' over links 810a and 810a' to the second node 804. The second node 804 then coherently combines the two digital communications 807b and 807b' with only a small risk of propagating errors. This scheme can be extended to three or more relay paths. In an one embodiment of the wireless relay based network 800, the relay stations 806a and 806b can be configured like example #4 and transmit high reliability symbols with higher power and low reliability symbols with smaller power in the digital communications 807b and 807b' which enables the second node 804 to combine them in a less risky manner since the erroneous symbols are likely to have a small reliability. Furthermore, it is likely in this embodiment that a less reliable portion of the received digital communication 807b from one relay station 806a (for example) would be matched by more reliable symbols in the digital communication 807b' from another relay station 806b (for example). In another embodiment of the wireless relay based network 800, the relay stations 806a and 806b know or can estimate the channel responses h0 ho, h1 or h1 on any of the links 808a, 808a', 810a and 810a' but on at least the second links 810b and 810b'. As such, one relay station 806a to construct a digital communication 807b that can be coherently combined at the second node 804 with a similar digital communication 807b' constructed by another relay station 806b. This is like the traditional wireless relay based network 500 shown in FIGURE 5 except that the relay stations 806a and 806b perform better using the advanced signal processing of the present invention. It should be appreciated that the relay station 806a (for example) can estimate the channel coefficient ho on the first link 808a if a known pilot symbol is transmitted together with the data in the digital communication 807a to relay station 806a. In this case, the channel coefficient ho can then be estimated by comparing the received corrupted pilot with the known pilot signal. The channel coefficient ho is then the same also for the data. In contrast, the relay station 806a (for example) has to know the channel coefficient h1 in advance of the transmission of the digital communication 807b to the second node 804. This is possible if the channel coefficient h1 is sufficiently slowly changing and there is some form of feedback between the relay station 806a and the second node 804. It could be that the second node 806a measures the channel coefficient h1 using the pilot method described above and then sends back information about h1 to relay station 806a with the hope that it will be the same (or close to) when it is time to transmit the next digital communication 807b. Another possibility is that there is also a link (not shown) from the second node 804 to the relay station 806a, and that both links use the same frequency band separated in time (TDD=time division duplex). In this case, any traffic in the direction from the second node 804 to relay station 806a experiences the same or almost the same channel coefficient h1 as the traffic in the direction from the relay station 806a to the second node 804. Hence, the relay station 806a can estimate the channel coefficient h1 in the direction from the second node 804 to relay station 806a and assume it to be the same in the other direction. Referring to FIGURE 9, there is shown a block diagram of yet another embodiment of a wireless relay based network 900 in accordance with the present invention. The wireless relay based network 900 includes a first node 902 that transmits information in coded and modulated digital communications to a second node 904 via two relay stations 906a and 906b. As shown, the first node 902 respectively transmits two channel coded and modulated signals 907a and 907a' (S0(t)) over two links 908a and 908a' to two relay stations 906a and 906b. The relay stations 906a and 906b then transmit coded and modulated digital communications 907b and 907b' (S1(t) and S1(t)) over two links 910a and 910a' to the second node 904. As can be seen, the second links 910a and 910a' have a larger bandwidth (higher bit rate) and may use a different air interface protocol than the first links 908a and 908a' in order to carry quantized reliability information about the code or information symbols and possibly channel amplitude and phase information from the relay stations 906a and 906b to the second node 904. This configuration is especially useful if the second node 904 is an MS because then the relay stations 906a and 906b can be seen as a wirelessly connected antenna system which enhances the reception at the second node 904. Another advantage of this set-up is that the antennas at the relay stations 906a and 906b could be spaced further apart and still enable virtual antenna diversity at the second node 904 (MS 904). In addition, the relay stations 906a and 906b could be multiple antenna/MIMO systems so that a second node 904 (MS 904) which has a low complexity could make use of method 700 of the present invention where part of the signal processing can be done in the relay stations 906a and 906b and part of it in the second node 904 (MS 904). Referring to FIGURE 10, there is shown a block diagram of yet another embodiment of a wireless relay based network 1000 in accordance with the present invention. The wireless relay based network 1000 includes a first node 1002 that transmits information in coded and modulated digital communication 1007a and 1007b to a second node 1004 via one relay station 1006. If the second node 1004 is able to decode the digital communication 1007b which is generated and transmitted from the relay station 1006 then it sends an acknowledgement message (ACK)(not shown) to the first node 1002. And, if the second node 1004 is not able to decode the digital communication 1007b then it sends a negative acknowledgment message (NAK) 1009 to the first node 1002. If the NAK 1009 is transmitted, then the first node 1002 transmits another digital communication 1007a' which is processed by the relay station 1006 and transmitted as digital communication 1007b' to the second node 1004. The second node 1004 then attempts to combine the two versions of the digital communication 1007b and 1007b'. Thus, if only one relay path is available between the first node 1002 and the second node 1004 then there is still a way to combine several temporally differentiated versions of the digital communication 1007b and 1007b' assuming there is an automatic repeat request (ARQ) protocol, an incremental redundancy ARQ protocol or a similar protocol running over the relay station 1006. Referring to FIGURE 11, there is shown a block diagram of still yet another embodiment of a wireless relay based network 1100 in accordance with the present invention. The wireless relay based network 1100 includes a first node 1102 that transmits information in coded and modulated digital communications to a second node 1104 via multiple relay stations 1106 (four shown) each of which implements method 700. The second node 1104 then combines the different digital communications. It should be appreciated that the number of relay stations 1106 shown in this wireless based relay network 1100 has been selected for simplicity of illustration and that the number of relay stations 1106 and their configuration is not a limitation on the invention. This is also true for the other embodiments of the wireless based relay networks 600, 700, 800, 900 and 1000. It should be appreciated that the aforementioned embodiments of the present invention can be combined in various ways to enhance the performance of wireless based relay networks. Although several embodiments of the present invention have been illustrated in the accompanying Drawings and described in the foregoing Detailed Description, it should be understood that the invention is not limited to the embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the spirit of the invention as set forth and defined by the following claims. WE CLAIM: 1. A wireless relay based network (600, 800, 900, 1000 and 1100) comprising: a first node (602, 802, 902, 1002 and 1102); at least one relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106); and a second node (604, 804, 904, 1004 and 1104), characteR1zed by said first node communicates with said second node via said at least one relay station, wherein each relay station is capable of: receiving (702) a digital communication (607a, 807a, 807a', 907a, 907a', 1007a and 1007a') from said first node; computing (704) a plurality of reliability values for a plurality of symbols in the received digital communication; and transmitting (706) a digital communication (607b, 807b, 807b', 907b, 907b', 1007b and 1007b') that has the computed reliability values embedded therein to said second node. 2. The wireless relay based network as claimed in Claim 1, wherein each relay station performs the computing step using a maximum a posteriori (MAP) filter (612a) that computes reliability values for code symbols based on a code structure of the received digital communication. 3. The wireless relay based network as claimed in Claim 2, wherein each MAP filter also filters the received digital communication and redistributes noise to unreliable parts in the transmitted digital communication. 4. The wireless relay based network as claimed in Claim 1, wherein each relay station performs the computing step using a soft output channel decoder (612b) that computes reliability values for information symbols based on a code structure of the received digital communication. 5. The wireless relay based network as claimed in Claim 4, wherein said soft output channel decoder employs: a maximum a posteriori (MAP) algoR1thm; a soft output Viterbi algoR1thm (SOVA); a Log-MAP algorithm; or a Max-LOG-MAP algoR1thm. 6. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication such that high reliability symbols are transmitted with higher power and low reliability symbols are transmitted with lower power to said second node. 7. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication in a manner where the reliability symbols are used to modulate an amplitude of the digital communication transmitted to said second node. 8. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication in a manner where the reliability symbols are used to modulate a phase of the digital communication transmitted to said second node. 9. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication in a manner where the reliability symbols are used to vary a bandwidth of the digital communication transmitted to said second node. 10. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication in a manner where the reliability symbols are used to vary a signal time occupation of the digital communication transmitted to said second node. 11. The wireless relay based network as claimed in Claim 1, wherein the computed reliability values are embedded in the transmitted digital communication in a manner where the reliability symbols are used to vary a signal constellation size of the digital communication transmitted to said second node. 12. The wireless relay based network as claimed in Claim 1, wherein said first node is: a base station; a mobile station; or a relay station. 13. The wireless relay based network as claimed in Claim 1, wherein said second node is: a base station; a mobile station; or a relay station. 14. The wireless relay based network as claimed in Claim 1, wherein each relay station is: a base station; a mobile station; or a stand alone relay station. 15. The wireless relay based network as claimed in Claim 1, wherein said received digital communication is: an uplink received digital communication; a downlink received digital communication; a base station peer-to-peer received digital communication; or a mobile station peer-to-peer received digital communication. 16. The wireless relay based network as claimed in Claim 1, wherein said transmitted digital communication is: an uplink transmitted digital communication; a downlink transmitted digital communication; a base station peer-to-peer transmitted digital communication; or 17. The wireless relay based network as claimed in Claim 1, wherein said second node combines the transmitted digital communications when each of the multiple relay stations transmit the digital communication. 18. The wireless relay based network as claimed in Claim 1, wherein said second node combines the transmitted digital communications when one relay station transmits multiple digital communications at different times. 19. The wireless relay based network as claimed in Claim 1, wherein one relay station knowing a channel response of a link between that relay station and said second node is able to construct a transmitted digital communication which is coherently combined at said second node with a similar transmitted digital communication received from another relay station. a mobile station peer-to-peer transmitted digital communication. 20. A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) capable of enabling communications between a first node (602, 802, 902, 1002 and 1102) and a second node (604, 804, 904, 1004 and 1104) by performing the following steps characteR1zed by: receiving (702) a coded/modulated digital communication (607a, 807a, 807a', 907a, 907a', 1007a and 1007a') from said first node; computing (704) a plurality of reliability values for a plurality of symbols in the received coded/modulated digital communication; and transmitting (706) a coded/modulated digital communication (607b, 807b, 807b', 907b, 907b', 1007b and 1007b') that has the computed reliability values embedded therein to said second node. 21. The relay station as claimed in Claim 20, further compR1sing a maximum a posteR1oR1 (MAP) filter (612a) that computes reliability values for code symbols based on a code structure of the received coded/modulated digital communication. 22. The relay station as claimed in Claim 20, further comprising a soft output channel decoder (612b) that computes reliability values for information symbols based on a code structure of the received coded/modulated digital communication. 23. The relay station as claimed in Claim 20, wherein the computed reliability values are explicitly embedded in the coded/modulated digital communication transmitted to said second node. 24. The relay station as claimed in Claim 20, wherein the computed reliability values are implicitly embedded in the coded/modulated digital communication transmitted to said second node. 25. The relay station as claimed in Claim 20, wherein said relay station is used in a wireless multi-hop network. 26. The relay station as claimed in Claim 20, wherein a link between said relay station and said first node has a smaller bandwidth than a link between said relay station and said second node. 27. The relay station as claimed in Claim 20, wherein each relay station is: a base station; a mobile station; or a stand alone relay station. 28. A method (700) for enabling a relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) to provide reliable digital communications between a first node (602, 802, 902, 1002 and 1102) and a second node (604, 804, 904, 1004 and 1104), said method characteR1zed by the following steps: receiving (702), at said relay station, a digital communication (607a, 807a, 807a', 907a, 907a', 1007a and 1007a') from said first node; computing (704), at said relay station, a plurality of reliability values for a plurality of symbols in the received digital communication; and transmitting (706), at said relay station, a digital communication (607b, 807b, 807b', 907b, 907b', 1007b and 1007b') that has the computed reliability values embedded therein to said second node. 29. The method as claimed in Claim 28, wherein said computing step if performed a maximum a posteriori (MAP) filter (612a) that computes reliability values for code symbols based on a code structure of the received digital communication. 30. The method as claimed in Claim 28, wherein said computing step if performed a soft output channel decoder (612b) that computes reliability values for information symbols based on a code structure of the received digital communication. 31. The method as claimed in Claim 28, wherein the computed reliability values are explicitly embedded in the digital communication transmitted to said second node. 32. The method as claimed in Claim 28, wherein the computed reliability values are implicitly embedded in the digital communication transmitted to said second node. A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) and a method (700) are described herein that enables reliable digital communications to occur between two nodes in a wireless relay based network (600, 800, 900, 1000 and 1100). The wireless relay based network includes a first node (602, 802, 902, 1002 and 1102) that transmits information in coded/modulated digital communications to a second node (604, 804, 904, 1004 and 1104) via one or more relay stations. And, each relay station is capable of: (1) receiving (702) a coded/modulated digital communication from the first node; (2) computing (704) a plurality of reliability values for a plurality of information symbols or coded symbols in the received coded/modulated digital communication; and (3) transmitting (706) a coded/modulated digital communication that has the computed reliability values embedded therein to the second node.

Full Text

BACKGROUND OF THE INVENTION
Field of the Invention
The present invention relates in general to the telecommunications field and, in
particular, to a relay station and method for enabling reliable digital communications to occur
between two nodes in a wireless relay based network.
Description of Related Art
Manufacturers and operators of wireless relay based networks are constantly trying to
develop new ways to improve the reliability of digital communications transmitted between
two nodes. The traditional wireless relay based networks described below each include one
station (node A) that transmits information in coded and modulated digital communications to
another station (node B) via one or more relay stations (RSs). The relay station can be part of
a base station (BS), a mobile station (MS) or a stand-alone relay station. The nodes A and B
can be a BS, a MS and/or a relay station. And, the digital communications could be either
uplink communications (link from MS to BS), downlink communications (link from BS to
MS), MS to MS communications or BS to BS communications. Examples of different
traditional wireless relay based networks and some of their respective drawbacks are briefly
discussed below with respect to FIGURES 1-5.
Referring to FIGURE 1 (PRIOR ART), there is shown a block diagram of a traditional
wireless relay based network 100. The wireless relay based network 100 includes a first node
102 (node A) that transmits information in coded and modulated digital communications to a
second node 104 (node B) via one relay station 106. As shown, the first node 102 transmits a
channel coded and modulated signal So(t) over a first link 108 that has a complex channel
coefficient h0 such that So(t) is received at relay station 106 as R1(t)=h0*S0(t)+n1(t), where
ni(t) is a noise sequence. The relay station 106 is a repeater that generate a signal S1(t) which
is an amplified version of the received sequence and is shown as S1(t)=A*R1(t). The relay
station 106 then transmits the generated signal Si(t) over a second link 110 that has a complex
channel coefficient h1 such that S1(t) is received at the second node 104 as

R2(t)=h1*S1(t)+n2(t), where n2(t) is a noise sequence. There are several drawbacks associated
with this type of wireless relay based network 100. First, the amplification of R1(t) at relay
station 106 amplifies not only the signal R1(t) but also the noise n1(t) caused by the first link
108. Secondly, the relay station 106 is not very efficient when it amplifies R1(t) since there is
a large amount of redundancy in So(t) due to channel coding.
Referring to FIGURE 2 (PRIOR ART), there is shown a block diagram of another
traditional wireless relay based network 200. The wireless relay based network 200 includes
a first node 202 (node A) that transmits information in coded and modulated digital
communications to a second node 204 (node B) via one relay station 206. As shown, the first
node 202 transmits a channel coded and modulated signal So(t) over a first link 208 that has a
complex channel coefficient ho such that So(t) is received at relay station 206 as
R1(t)=h0*So(t)+n1(t), where ni(t) is a noise sequence. The relay station 206 then decodes, re-
encodes, re-modulates and transmits a signal S1(t)=Al*Soest(t) over a second link 210 that has
a complex channel coefficient h1 which is received at the second node 204. This is a good
solution whenever the relay station 206 makes a correct decision when it estimates and re-
encodes So(t). However, when the relay station 206 makes an incorrect decision, the second
link 210 further propagates the information errors and increases the bit error rate of the signal
Si(t) transmitted to second node 204.
Referring to FIGURE 3 (PRIOR ART), there is shown a block diagram of yet another
traditional wireless relay based network 300. The wireless relay based network 300 includes
a first node 302 (node A) that transmits information in coded and modulated digital
communications to a second node 304 (node B) via one relay station 306. As shown, the first
node 302 transmits a channel coded and modulated signal So(t) over a first link 308 that has a
complex channel coefficient ho such that So(t) is received at relay station 306 as
R1(t)=ho*So(t)+n1(t), where n1(t) is a noise sequence. The relay station 306 then decodes, re-
encodes, re-modulates, amplifies and transmits a signal S1(t) over a second link 310 that has a
complex channel coefficient h1 which is received at the second node 304. In this example, the
relay station 306 checks the correctness of the decoding of So(t) using for example a cyclic
redundancy check (CRC) and only re-generates, re-modulates and transmits S1(t)=Al*So(t) in
case of correctness, otherwise the relay station 306 simply amplifies and retransmits the

received signal as S1(t)=A2*R1(t). This solution is problematic since the relay station 306
needs to make a hard decision on the information symbols in S1(t) instead of having the
second node 304 make that hard decision. It is well known that in a communication chain it is
advantageous not to make hard decisions until late as possible in the chain. As described
below in detail there are some known ways to avoid making hard decisions in relay stations
which include using a second relay path or using an automatic repeat request (ARQ) protocol.
Referring to FIGURE 4 (PRIOR ART), there is shown a block diagram of yet another
traditional wireless relay based network 400. The wireless relay based network 400 includes
a first node 402 (node A) that transmits information in coded and modulated digital
communications to a second node 404 (node B) via one relay station 406. As shown, the first
node 402 transmits a channel coded and modulated signal So(t) over a first link 408 that has a
complex channel coefficient ho to relay station 406. The relay station 406 then transmits
S1(t)=Q(R1(t)) which is a quantized base band representation of R1(t) over a second link 410
that has a complex channel coefficient h1 to the second node 404. In this solution, the second
link 410 generally has a larger bandwidth and uses a different air interface protocol than the
first link 408. This solution is problematic in that the quantized base band signal S1(t)
contains a large amount of data about phase and noise that does not necessarily need to be
sent to the second node 404.
Referring to FIGURE 5 (PRIOR ART), there is shown a block diagram of a traditional
wireless relay based network 500. The wireless relay based network 500 includes a first node
502 (node A) that transmits information in coded and modulated digital communications to a
second node 504 (node B) via two relay stations 506a and 506b. As shown, the first node 502
transmits a channel coded and modulated signal So(t) over a link 508a that has a complex
channel coefficient ho to relay station 506a. At the same time, the first node 502 also
transmits So(t) over a link 508b that has a complex channel coefficient ho to relay station
506b. The relay stations 506a and 506b know or can estimate their respective channel
coefficients ho, ho, h1 and h1 on links 508a, 508b, 510a and 510b. As such, relay station 506a
can transmit S1(t)=A*conj(h0)*conj(h1)*R1(t) over link 510a to the second node 504. And,
relay station 506b can transmit S1(t)=A*conj(h0)*conj(h1)*R1(t) over link 510b to the second
node 504. As is well known the signals R1(t) and R1(t) because of the scaling and phase

shifts involved can be added coherently in a maximum ratio combining sense when they are
received at the second node 504. This scheme can be extended to three or more relay paths.
Even with multiple relay paths there is still a reliability problem since each relay station 506a
and 506b may make a different decision in decoding signals R1(t) and R1(t) which makes it
difficult for the second node 504 to coherently combine signals S1(t) and S1(t). One possible
solution when there are multiple relay paths is for each relay station 506a and 506b to check
an error detecting outer code in signals R1(t) and R1(t). If either R1(t) and/or R1(t) checks out
OK then the respective relay station 506a and/or 506b transmits the corresponding S1(t)
and/or S1(t), otherwise the respective relay station 506a and/or 506b does not send anything
or they can send an amplified version of the respective received signal(s). The problem with
this solution is that it relies on the following assumptions that (1) at least one relay station
506a or 506b has correctly decoded signal R1(t) and R1(t) and (2) that the corresponding
second link 510a or 510b for such relay station 506a or 506b is sufficiently good to enable the
decoding of signals S1(t) or S1(t). Accordingly, there is a need for a new signal processing
solution that can be implemented in a relay station which addresses the aforementioned
shortcomings and other shortcomings of the traditional wireless relay based networks 100,
200, 300, 400 and 500. This need and other needs are satisfied by the wireless relay based
network and relay station of the present invention.
BR1EF DESCR1PTION OF THE INVENTION
The present invention includes a relay station and a method that enables reliable
digital communications to occur between two nodes in a wireless relay based network. The
wireless relay based network includes a first node that transmits information in
coded/modulated digital communications to a second node via one or more relay stations.
And, each relay station is capable of: (1) receiving a coded/modulated digital communication
from the first node; (2) computing a plurality of reliability values for a plurality of
information symbols or coded symbols in the received coded/modulated digital
communication; and (3) transmitting a coded/modulated digital communication that has the
computed reliability values embedded therein to the second node.

BR1EF DESCR1PTION OF THE DRAWINGS
A more complete understanding of the present invention may be had by reference to
the following detailed descR1ption when taken in conjunction with the accompanying drawings
wherein:
FIGURES 1-5 (PR1OR ART) are block diagrams of five different traditional wireless
relay based networks;
FIGURE 6 is a block diagram of a wireless relay based network and a relay station in
accordance with one embodiment of the present invention;
FIGURE 7A is a flowchart illustrating the steps of a preferred method for using the
relay station shown in FIGURE 6 in accordance with the present invention;
FIGURE 7B is a diagram illustrating how a Log-APP filter can implement an APP
algoR1thm which is one way to help enable the method shown in FIGURE 7A;
FIGURE 8 is a block diagram of a wireless relay based network in accordance with
another embodiment of the present invention;
FIGURE 9 is a block diagram of a wireless relay based network in accordance with
yet another embodiment of the present invention;
FIGURE 10 is a block diagram of a wireless relay based network in accordance with
still yet another embodiment of the present invention; and
FIGURE 11 is a block diagram of a wireless relay based network in accordance with
yet another embodiment of the present invention.
DETAILED DESCR1PTION OF THE DRAWINGS
ReferR1ng to FIGURES 6 and 7A, there are disclosed preferred embodiments of a
wireless relay based network 600 and method 700 in accordance with the present invention.

The wireless relay based network 600 includes a first node 602 (node A) that transmits
information in coded and modulated digital communications to a second node 604 (node B)
via a relay station 606. Like the traditional wireless relay based networks 100, 200, 300, 400
and 500 shown in FIGURES 1-5, the relay station 606 can be part of a BS, MS or a stand-
alone relay station. The first node 602 and the second node 604 can be a BS, a MS and/or a
relay station. And, the digital communications could be either uplink communications (link
from MS to BS), downlink communications (link from BS to MS), MS to MS
communications or BS to BS communications. However, the relay station 606 of the present
invention has a functionality shown in method 700 that is a marked improvement over the
functionalities of the traditional relay stations 106, 206, 306, 406 and 506 descR1bed above
with respect to the traditional wireless relay based networks 100, 200, 300, 400 and 500.
The relay station 606 basically enables information to be reliably transmitted between
the first node 602 and the second node 604 by receiving (step 702) a coded and modulated
digital communication 607a which was transmitted from the first node 602 over a wireless
link 608 that has a complex channel coefficient h0 The relay station 606 then computes (step
704) a plurality of reliability values for a plurality of symbols in the received digital
communication 607a (R1(t)). There are many ways of which two are descR1bed in detail
below as to how the relay station 606 can compute the reliability values for the symbols in the
received digital communication 607a.
In the first way, the relay station 606 utilizes a maximum a posteoR1 (MAP) filter
612a to compute the reliability values for code symbols based on a code structure of the
received coded digital communication 607a. For more details about the MAP filter 612a
reference is made to the following articles all of which are incorporated by reference herein:
• J. Lodge et al. "Separable MAP "filters" for the Decoding of Product and
Concatenated Codes" Proceedings of IEEE International Conference on
Communication, pp 1740-1745, 1993.*
• I. Land et al. "On the Interpretation of the APP AlgoR1thm as an LLR Filter"
ISIT 2000, Sorrento, Italy, June 25-30, 2000.*

* It should be noted that in the literature the acronym MAP (maximum a posteR1oR1) can be
called APP (a posteR1oR1 probability, Log-APP or Max-log-APP.
FIGURE 7B is a diagram showing how a Log-APP filter 612a can implement an APP
algoR1thm that helps perform steps 702, 704 and 706 in the method 700. It should be
appreciated that the notation used herein is the same notation as used in Land et al. where the
log-likelihood values for the code symbols are denoted by L(X) and the log-likelihood values
for the information symbols are denoted by L(U) with superscR1pts + or - depending if the
corresponding reliabilities are considered before (-) or after (+) the Log-APP filter 612a.
The log-likelihood value L(X) for a binary code symbol X is In (P(X=+1 | received
data) / P(X=-1 I received data))(see Land et al.) As such, it follows that equally probable
symbols then have log-likelihood value (L-value) of zero and a highly likely '+1' have a large
positive L-value, and conversely a highly likely '-1' have a large negative L-value. Most
often the information symbols are equally likely from the receivers perspective before
processing, hence L(U")=0 is often assumed which further simplifies the calculations in the
Log-APP filter 612a. After processing the updated L(X+) and/or L(U+) these values can be
embedded into transmitted radio signal (see step 706 in FIGURE 7A). In the most basic and
preferred form either information or code symbol reliabilities are embedded, but of course
they can be combined as well. If only code symbol reliabilities are to be used, of course the
reliability calculator (Log-APP in the example) need not calculate information reliabilities or
vice versa. FIGURE 7B shows the case where all information is used in the most general
form.
The Log-APP filter 612a can be interpreted as using the code structure to 'amplify' the
reliability and the more amplification there is the more reliable the symbols are up to a limit
decided by the code parameters. If only L(X") and L(X+) are used then the Log-APP filter
612a can be seen as a very non-linear log likelihood filter which outputs values in the same
domain with the same type of values and code symbols as the output but with an improved
signal to noise ratio. It should be appreciated that since the input and outputs are of the same
form (L(U) and/or L(X)) then in some cases, after a first application of the APP algoR1thm that

the output L(X+) can be connected to the input L(X-) and the APP algoR1thm can be run again
with these new values as input. The same follows for L(U)-values.
An additional functionality and advantage of using the Log-APP filter 612a shown in
FIGURE 7B is that it makes use of the redundancy and structure of the channel code in the
received digital communication 607a to clean-up or filter that digital communication 607a and
redistR1bute noise to unreliable parts in the digital communication 607b (S1(t)) transmitted to
the second node 604. It should be appreciated that the Log-App filter 612a does not
necessaR1ly regenerate a valid code word but computes a code symbol by code symbol
reliability value.
In the second way, the relay station 606 utilizes a soft output channel decoder 612b to
compute the reliability values for information symbols based on a code structure of the
received coded digital communication 607a. To accomplish this, the soft output channel
decoder 612b can employ for example: (1) a maximum a posteR1oR1 (MAP) algorithm; (2) a
soft output Viterbi algoR1thm (SOVA); (3) a Log-MAP algoR1thm; or (4) a Max-LOG-MAP
algoR1thm. For more details about the soft output channel decoders 612b reference is made to
the following articles all of which are incorporated by reference herein:
• P. Robertson et al. "Optimal and Sub-Optimal Maximum a Posteriori Algorithms
Suitable for Turbo Decoding" European Transaction on Telecommunications (ETT),
Vol. 8, No. 2, pp. 1-16, 1997.
• J. Hagenauer et al. "A Viterbi Algorithm with Soft-Decision Outputs and its
Applications" IEEE Globecom, pp. 1680-1686, 1989.
The reliability values that are computed for the code symbols or information symbols
could have anyone of a wide variety of formats. For example, a binary '1' symbol could have
a reliability value +1.0 if there is a 100% certainty of a T. And, a binary '0' symbol could
have a reliability value -1.0 if there is a 100% certainty of a binary '0'. Moreover, the
reliability value can be 0.0 if the bit is completely unreliable (T and '0' equally probable). In
another example, a reliability value of+0.2 could indicate that a binary T symbol is more

likely to be correct but a +0.8 would indicate an even stronger conviction of a binary '1'
symbol. It should be appreciated that there could be more possible reliability values than
symbols in order to be able to exploit 'soft' information. For example, a reliability value of
{+1.0, 0, -1.0} could denote three possible reliabilities for two symbols '0' and T. In another
example, a reliability value of {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} could denote seven
possible reliability values for '0' and '1' symbols and also provide a better resolution. The
seven possible reliability values could be represented by a 3-bit code.
After the relay station 606 computes reliability values for either the code symbols or
information symbols within the received digital communication 607a, the relay station 606
then transmits (step 706) a coded and modulated digital communication 607b that has the
computed reliability values embedded therein over a wireless link 614 which has a complex
channel coefficient h1 to the second node 604. There are a vaR1ety of ways for embedding the
reliability values in the transmitted digital communication 607b (S1(t)). For example, the
reliability values can be explicitly embedded into the transmitted coded digital
communication 607b by actually transmitting data representing those reliability values. In
another example, the reliability values can be implicitly embedded into the transmitted digital
communication 607b by using the reliability values to modulate the amplitude (and/or phase)
of the transmitted digital communication 607b. Several different examples are descR1bed
below on how the reliability values can be embedded into the transmitted digital
communication 607b.
Example #1: The reliability values can actually be embedded in the transmitted digital
communication 607b if one increases the bandwidth or bitrate of the transmitted digital
communication 607b. And, instead of transmitting reliability values of'0' or '1' which takes a
single bit one could get a better resolution by tripling the bit rate and transmitting three bits
denoting one of seven reliability values {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} for each of the
symbols in the digital communication 607b. In this example, the relay station 606 would act
as a supportive processing element for the second node 604 (e.g., user terminal). This option
is especially attractive if the wireless relay based network has a very high capacity second link
610 like the ones shown in FIGURE 9.

Example #2: If multiple reliability values {+1.0, +0.5, +0.2, 0, -0.2, -0.5, -1.0} for each of the
symbols are embedded in the transmitted digital communication 607b, then something has to
be increased like the bandwidth as descR1bed in example #1. Another option is to increase the
size of the signal constellation from 2 to 8 (for example) to embed multiple reliability values
associated with each symbol in the transmitted digital communication 607b. For example, a
binary phase shift key (BPSK) may be used when two reliability values for each symbol are
embedded in the transmitted digital communication 607b. If there are seven reliability values
as described in example #1 then 8-PSK is an option.
Example #3: To implicitly transmit reliability values one can use different signal constellation
sizes (modulation schemes) for symbols with different reliabilities such that more energy per
symbol is spent on the reliable symbols in the transmitted digital communication 607b.
Example #4: One could generate the transmitted digital communication 607b and modulate it
using the reliability values for the corresponding bits in the digital communication 607b. For
example, one can amplitude modulate the transmitted digital communication 607b with the
reliability of symbol at time t, REL(t) and transmit the coded digital communication 607b
(e.g., S2(t)=REL(t)*Si(t)). In this case, the reliable parts of the digital communication 607b
are transmitted with higher power than the low reliability parts of the digital communication
607b. This is done because there is no reason to waste energy on already unreliable bits since
the reliability of a bit can only decrease in the transmission chain.
Example #5: The reliable parts of the digital communication 607b could use a longer signal
time occupation signal than the one for unreliable parts. This is another way of ensuring that
reliable parts are transmitted with higher total energy. An advantage of this example when
compared to example #4 is that the amplitude of the transmitted digital communication 607b
is now constant which is sometimes desirable. It should be appreciated that any modulation
scheme (amplitude modulation, phase modulation) implies an expansion of the bandwidth on
the link 610 to the second node 604.

Again, it should be appreciated that there are other ways that are known or can be
subsequently developed which can be used to embed reliability values into the transmitted
digital communication 607b in accordance with the present invention.
Referring to FIGURE 8, there is shown a block diagram of another embodiment of a
wireless relay based network 800 in accordance with the present invention. The wireless
relay based network 800 includes a first node 802 that transmits information in coded and
modulated digital communications to a second node 804 via two relay stations 806a and 806b.
As shown, the first node 802 transmits a channel coded and modulated signal 807a (So(t))
over link 808a that has a complex channel coefficient ho to relay station 806a. At the same
time, the first node 802 also transmits digital communication 808a' (So(t)) over a link 808a'
that has a complex channel coefficient ho to relay station 806b. The relay stations 806a and
806b respectively perform the signal processing associated with method 700 and transmit
digital communications 807b and 807b' over links 810a and 810a' to the second node 804.
The second node 804 then coherently combines the two digital communications 807b and
807b' with only a small risk of propagating errors. This scheme can be extended to three or
more relay paths.
In an one embodiment of the wireless relay based network 800, the relay stations 806a
and 806b can be configured like example #4 and transmit high reliability symbols with higher
power and low reliability symbols with smaller power in the digital communications 807b and
807b' which enables the second node 804 to combine them in a less risky manner since the
erroneous symbols are likely to have a small reliability. Furthermore, it is likely in this
embodiment that a less reliable portion of the received digital communication 807b from one
relay station 806a (for example) would be matched by more reliable symbols in the digital
communication 807b' from another relay station 806b (for example).
In another embodiment of the wireless relay based network 800, the relay stations
806a and 806b know or can estimate the channel responses h0 ho, h1 or h1 on any of the links
808a, 808a', 810a and 810a' but on at least the second links 810b and 810b'. As such, one
relay station 806a to construct a digital communication 807b that can be coherently combined

at the second node 804 with a similar digital communication 807b' constructed by another
relay station 806b. This is like the traditional wireless relay based network 500 shown in
FIGURE 5 except that the relay stations 806a and 806b perform better using the advanced
signal processing of the present invention.
It should be appreciated that the relay station 806a (for example) can estimate the
channel coefficient ho on the first link 808a if a known pilot symbol is transmitted together
with the data in the digital communication 807a to relay station 806a. In this case, the
channel coefficient ho can then be estimated by comparing the received corrupted pilot with
the known pilot signal. The channel coefficient ho is then the same also for the data. In
contrast, the relay station 806a (for example) has to know the channel coefficient h1 in
advance of the transmission of the digital communication 807b to the second node 804. This
is possible if the channel coefficient h1 is sufficiently slowly changing and there is some form
of feedback between the relay station 806a and the second node 804. It could be that the
second node 806a measures the channel coefficient h1 using the pilot method described above
and then sends back information about h1 to relay station 806a with the hope that it will be the
same (or close to) when it is time to transmit the next digital communication 807b. Another
possibility is that there is also a link (not shown) from the second node 804 to the relay station
806a, and that both links use the same frequency band separated in time (TDD=time division
duplex). In this case, any traffic in the direction from the second node 804 to relay station
806a experiences the same or almost the same channel coefficient h1 as the traffic in the
direction from the relay station 806a to the second node 804. Hence, the relay station 806a
can estimate the channel coefficient h1 in the direction from the second node 804 to relay
station 806a and assume it to be the same in the other direction.
Referring to FIGURE 9, there is shown a block diagram of yet another embodiment of
a wireless relay based network 900 in accordance with the present invention. The wireless
relay based network 900 includes a first node 902 that transmits information in coded and
modulated digital communications to a second node 904 via two relay stations 906a and 906b.
As shown, the first node 902 respectively transmits two channel coded and modulated signals
907a and 907a' (S0(t)) over two links 908a and 908a' to two relay stations 906a and 906b. The
relay stations 906a and 906b then transmit coded and modulated digital communications 907b

and 907b' (S1(t) and S1(t)) over two links 910a and 910a' to the second node 904. As can be
seen, the second links 910a and 910a' have a larger bandwidth (higher bit rate) and may use a
different air interface protocol than the first links 908a and 908a' in order to carry quantized
reliability information about the code or information symbols and possibly channel amplitude
and phase information from the relay stations 906a and 906b to the second node 904. This
configuration is especially useful if the second node 904 is an MS because then the relay
stations 906a and 906b can be seen as a wirelessly connected antenna system which enhances
the reception at the second node 904. Another advantage of this set-up is that the antennas at
the relay stations 906a and 906b could be spaced further apart and still enable virtual antenna
diversity at the second node 904 (MS 904). In addition, the relay stations 906a and 906b
could be multiple antenna/MIMO systems so that a second node 904 (MS 904) which has a
low complexity could make use of method 700 of the present invention where part of the
signal processing can be done in the relay stations 906a and 906b and part of it in the second
node 904 (MS 904).
Referring to FIGURE 10, there is shown a block diagram of yet another embodiment
of a wireless relay based network 1000 in accordance with the present invention. The
wireless relay based network 1000 includes a first node 1002 that transmits information in
coded and modulated digital communication 1007a and 1007b to a second node 1004 via one
relay station 1006. If the second node 1004 is able to decode the digital communication
1007b which is generated and transmitted from the relay station 1006 then it sends an
acknowledgement message (ACK)(not shown) to the first node 1002. And, if the second
node 1004 is not able to decode the digital communication 1007b then it sends a negative
acknowledgment message (NAK) 1009 to the first node 1002. If the NAK 1009 is
transmitted, then the first node 1002 transmits another digital communication 1007a' which is
processed by the relay station 1006 and transmitted as digital communication 1007b' to the
second node 1004. The second node 1004 then attempts to combine the two versions of the
digital communication 1007b and 1007b'. Thus, if only one relay path is available between
the first node 1002 and the second node 1004 then there is still a way to combine several
temporally differentiated versions of the digital communication 1007b and 1007b' assuming
there is an automatic repeat request (ARQ) protocol, an incremental redundancy ARQ
protocol or a similar protocol running over the relay station 1006.

Referring to FIGURE 11, there is shown a block diagram of still yet another
embodiment of a wireless relay based network 1100 in accordance with the present invention.
The wireless relay based network 1100 includes a first node 1102 that transmits information
in coded and modulated digital communications to a second node 1104 via multiple relay
stations 1106 (four shown) each of which implements method 700. The second node 1104
then combines the different digital communications. It should be appreciated that the number
of relay stations 1106 shown in this wireless based relay network 1100 has been selected for
simplicity of illustration and that the number of relay stations 1106 and their configuration is
not a limitation on the invention. This is also true for the other embodiments of the wireless
based relay networks 600, 700, 800, 900 and 1000.
It should be appreciated that the aforementioned embodiments of the present invention
can be combined in various ways to enhance the performance of wireless based relay
networks.
Although several embodiments of the present invention have been illustrated in the
accompanying Drawings and described in the foregoing Detailed Description, it should be
understood that the invention is not limited to the embodiments disclosed, but is capable of
numerous rearrangements, modifications and substitutions without departing from the spirit of
the invention as set forth and defined by the following claims.

WE CLAIM:
1. A wireless relay based network (600, 800, 900, 1000 and 1100)
comprising:
a first node (602, 802, 902, 1002 and 1102);
at least one relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106); and
a second node (604, 804, 904, 1004 and 1104), characteR1zed by said first node
communicates with said second node via said at least one relay station, wherein each
relay station is capable of:
receiving (702) a digital communication (607a, 807a, 807a', 907a, 907a',
1007a and 1007a') from said first node;
computing (704) a plurality of reliability values for a plurality of symbols in
the received digital communication; and
transmitting (706) a digital communication (607b, 807b, 807b', 907b,
907b', 1007b and 1007b') that has the computed reliability values embedded
therein to said second node.
2. The wireless relay based network as claimed in Claim 1, wherein each
relay station performs the computing step using a maximum a posteriori (MAP) filter
(612a) that computes reliability values for code symbols based on a code structure of the
received digital communication.
3. The wireless relay based network as claimed in Claim 2, wherein each
MAP filter also filters the received digital communication and redistributes noise to
unreliable parts in the transmitted digital communication.
4. The wireless relay based network as claimed in Claim 1, wherein each
relay station performs the computing step using a soft output channel decoder (612b)
that computes reliability values for information symbols based on a code structure of the
received digital communication.

5. The wireless relay based network as claimed in Claim 4, wherein said soft
output channel decoder employs:
a maximum a posteriori (MAP) algoR1thm;
a soft output Viterbi algoR1thm (SOVA);
a Log-MAP algorithm; or
a Max-LOG-MAP algoR1thm.
6. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication such
that high reliability symbols are transmitted with higher power and low reliability symbols
are transmitted with lower power to said second node.
7. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication in a
manner where the reliability symbols are used to modulate an amplitude of the digital
communication transmitted to said second node.
8. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication in a
manner where the reliability symbols are used to modulate a phase of the digital
communication transmitted to said second node.
9. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication in a
manner where the reliability symbols are used to vary a bandwidth of the digital
communication transmitted to said second node.
10. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication in a
manner where the reliability symbols are used to vary a signal time occupation of the
digital communication transmitted to said second node.
11. The wireless relay based network as claimed in Claim 1, wherein the
computed reliability values are embedded in the transmitted digital communication in a

manner where the reliability symbols are used to vary a signal constellation size of the
digital communication transmitted to said second node.
12. The wireless relay based network as claimed in Claim 1, wherein said first
node is:
a base station;
a mobile station; or
a relay station.
13. The wireless relay based network as claimed in Claim 1, wherein said
second node is:
a base station;
a mobile station; or
a relay station.
14. The wireless relay based network as claimed in Claim 1, wherein each
relay station is:
a base station;
a mobile station; or
a stand alone relay station.
15. The wireless relay based network as claimed in Claim 1, wherein said
received digital communication is:
an uplink received digital communication;
a downlink received digital communication;
a base station peer-to-peer received digital communication; or
a mobile station peer-to-peer received digital communication.
16. The wireless relay based network as claimed in Claim 1, wherein said
transmitted digital communication is:
an uplink transmitted digital communication;
a downlink transmitted digital communication;
a base station peer-to-peer transmitted digital communication; or

17. The wireless relay based network as claimed in Claim 1, wherein said second node
combines the transmitted digital communications when each of the multiple relay
stations transmit the digital communication.
18. The wireless relay based network as claimed in Claim 1, wherein said second node
combines the transmitted digital communications when one relay station transmits
multiple digital communications at different times.
19. The wireless relay based network as claimed in Claim 1, wherein one relay station
knowing a channel response of a link between that relay station and said second node is
able to construct a transmitted digital communication which is coherently combined at
said second node with a similar transmitted digital communication received from another
relay station.
a mobile station peer-to-peer transmitted digital communication.
20. A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) capable of enabling
communications between a first node (602, 802, 902, 1002 and 1102) and a second
node (604, 804, 904, 1004 and 1104) by performing the following steps characteR1zed
by:
receiving (702) a coded/modulated digital communication (607a, 807a, 807a', 907a,
907a', 1007a and 1007a') from said first node;
computing (704) a plurality of reliability values for a plurality of symbols in the received
coded/modulated digital communication; and
transmitting (706) a coded/modulated digital communication (607b, 807b, 807b', 907b,
907b', 1007b and 1007b') that has the computed reliability values embedded therein to
said second node.
21. The relay station as claimed in Claim 20, further compR1sing a maximum a
posteR1oR1 (MAP) filter (612a) that computes reliability values for code symbols based on
a code structure of the received coded/modulated digital communication.

22. The relay station as claimed in Claim 20, further comprising a soft output
channel decoder (612b) that computes reliability values for information symbols based
on a code structure of the received coded/modulated digital communication.
23. The relay station as claimed in Claim 20, wherein the computed reliability
values are explicitly embedded in the coded/modulated digital communication
transmitted to said second node.
24. The relay station as claimed in Claim 20, wherein the computed reliability
values are implicitly embedded in the coded/modulated digital communication
transmitted to said second node.
25. The relay station as claimed in Claim 20, wherein said relay station is
used in a wireless multi-hop network.
26. The relay station as claimed in Claim 20, wherein a link between said
relay station and said first node has a smaller bandwidth than a link between said relay
station and said second node.
27. The relay station as claimed in Claim 20, wherein each relay station is:
a base station;
a mobile station; or
a stand alone relay station.
28. A method (700) for enabling a relay station (606, 806a, 806b, 906a, 906b,
1006 and 1106) to provide reliable digital communications between a first node (602,
802, 902, 1002 and 1102) and a second node (604, 804, 904, 1004 and 1104), said
method characteR1zed by the following steps:
receiving (702), at said relay station, a digital communication (607a, 807a, 807a',
907a, 907a', 1007a and 1007a') from said first node;
computing (704), at said relay station, a plurality of reliability values for a plurality
of symbols in the received digital communication; and

transmitting (706), at said relay station, a digital communication (607b, 807b,
807b', 907b, 907b', 1007b and 1007b') that has the computed reliability values
embedded therein to said second node.
29. The method as claimed in Claim 28, wherein said computing step if
performed a maximum a posteriori (MAP) filter (612a) that computes reliability values for
code symbols based on a code structure of the received digital communication.
30. The method as claimed in Claim 28, wherein said computing step if
performed a soft output channel decoder (612b) that computes reliability values for
information symbols based on a code structure of the received digital communication.
31. The method as claimed in Claim 28, wherein the computed reliability
values are explicitly embedded in the digital communication transmitted to said second
node.
32. The method as claimed in Claim 28, wherein the computed reliability
values are implicitly embedded in the digital communication transmitted to said second
node.

A relay station (606, 806a, 806b, 906a, 906b, 1006 and 1106) and a method (700) are
described herein that enables reliable digital communications to occur between two nodes in a
wireless relay based network (600, 800, 900, 1000 and 1100). The wireless relay based
network includes a first node (602, 802, 902, 1002 and 1102) that transmits information in
coded/modulated digital communications to a second node (604, 804, 904, 1004 and 1104)
via one or more relay stations. And, each relay station is capable of: (1) receiving (702) a
coded/modulated digital communication from the first node; (2) computing (704) a plurality
of reliability values for a plurality of information symbols or coded symbols in the received
coded/modulated digital communication; and (3) transmitting (706) a coded/modulated digital
communication that has the computed reliability values embedded therein to the second node.

RELAY STATION AND METHOD FOR ENABLING RELIABLE DIGITAL COMMUNICATIONS BETWEEN TWO NODES IN A WIRELES RELAY BASED NETWORK

Documents:

Inventors:

PCT Conventions: