Multi-Antenna Multicasting with Opportunistic Multicast Scheduling and Optimized Space-Time Transmission

多天線群播環境下之機會式群播排程和最佳化空時傳輸技術

方博雋(Po-Chun Fang)

指導教授 ： 洪樂文

國立清華大學/電機資訊學院/通訊工程研究所/碩士(2010年)

https://doi.org/10.6843/NTHU.2010.00415

摘要

在本篇論文中，我們提出實體層群播系統下之(physical layer multicasting)機會式群播排程(OMS)以及最佳化空時傳輸(OST)技術。我們考慮一個多天線下行鏈結(multi-antenna downlink)的環境，其中包含一個裝有多根天線的基地台以及一群各自僅有一根天線的用戶端。在此系統中，此基地台欲將相同的訊息傳送給指定的群體用戶。針對這樣的系統，其群播應用下的通道容量(channel capacity)是已知的，但要達到此容量則需相當的複雜度。在文獻中，多天線群播的方法大多侷限於空間多工(spatial multiplexing)和傳送端波束成形(transmit beamforming)的應用。為了提升效能，我們將最近提出的機會式群播排程加入到這些方法。多點傳輸排程中的訊息在來源(source)端利用噴泉碼(fountain code)進行編碼，並且在每個時間槽 (time slot)中只有最佳的一組用戶進入排程。利用極值定理(extreme value theory)，我們求出使用機會式群播排程的系統之平均吞吐量(average throughput)。這些表達式被分別用來求得使用群播排程的空間多工和傳送端波束成形的最佳用戶選擇比率(optimal user selection ratio)。
使用群播排程的空間多工和傳送端波束成型兩種方法對於通道輸入協方差矩陣(covariance matrix)都有加上嚴格的限制，並且在傳送端擁有通道狀態訊息(channel state information)的好處沒有被完全被利用。為了解決這個問題，我們進一步提出最佳化空時傳送的方法，其中通道輸入協方差矩陣的限制被移除了。我們所提出的最佳化空時傳送包含了傳送端波束成型和空間多工，並且提供更好的效能。由於整體上滿秩(full rank)的協方差矩陣對於最佳化空時傳送法的問題為最佳解，為了降低編碼複雜度，我們可以經由對矩陣的秩加上限制，也就是減少訊號傳輸空間維度(spatial dimension)的數目來達成。然而
，這會產生一個非凸面(non-convex)的最佳化問題。因此我們提出一個直覺上的維度選擇演算法(heuristic dimension selection algorithm)以避免解決複雜的非凸面最佳化問題。模擬的結果證實這個演算法比機會式群播排程的空間多工和傳送端波束成形較優異。為了降低最佳化空時傳送所需要的回饋(feedback)通道狀態訊息，我們提出一個減少通道回饋的方法。
最後，我們將這些概念延伸到用戶端具有多天線(MIMO)的環境，展現多天線最佳化空時傳送問題能經由凸面最佳化來解決。降低通道回饋的方法同樣的被延伸到多天線環境中。

關鍵字

群播 ； 機會式群播排程 ； 最佳化空時傳輸

並列摘要

Physical layer multicasting with opportunistic user selection and
optimized space-time transmission is examined in this work. We consider
a multi-antenna downlink scenario where a single base-station, equipped
with multiple antennas, transmits common information to a given set of
users. We first study the case where each user has a single receive antenna, that is, the multi-user MISO scenario. Although the multicast capacity is known, the complexity of achieving it is very high.
In previous works, pratical
multi-antenna multicasting solutions have been restricted to spatial
multiplexing and transmit beamforming. Here, we incorporate the recently
proposed opportunistic multicast scheduling (OMS) scheme into these techniques to improve the performance. The messages in OMS
are source coded using fountain codes and only an optimal subset of users is
scheduled at each time slot. Capitalizing on extreme value
theory, we derive analytical expressions for the average
throughput of OMS aided systems. These expressions are also utilized to obtain the optimal user selection
ratio for both OMS aided spatial multiplexing and transmit beamforming scenarios.

Both OMS aided spatial multiplexing and transmit beamforming impose stringent constraints on the channel input covariance matrix. And the benefits of having channel state information (CSI) at the transmitter is not fully exploited. To solve this problem, we further propose an optimized space-time transmission
(OST) scheme where the constraints on the channel input covariance matrix are removed. The proposed OST subsumes both transmit
beamforming and spatial multiplexing and, thus, outperforms both
conventional schemes. Since the OST problem is in general full rank optimal, we can reduce coding complexity by adding an rank constraint on the channel input covariance matrix. However, the optimization problem becomes non-convex. We then proposed an heuristic dimension selection algorithm which avoids solving the non-convex optimization problem. Simulation results verifies the superiority of this algorithm compared to the OMS aided spatial multiplexing and transmit beamforming scenarios. To relax the system overhead to feedback the full CSI needed by the OMS aided OST scheme, we also propose a method to reduce the channel feedback requirement.

Finally, these concepts are further extended to the MIMO scenario. We show that the MIMO OST problem can be solved via convex optimization with full CSI at the transmitter. The channel feedback reduction algorithm is also extended to the MIMO scenario.

並列關鍵字

multicast ； opportunistic multicast scheduling ； optimized space-time transmission

參考文獻

[1] U. C. Kozat, “On the throughput capacity of opportunistic multicasting with erasure

codes,” in Proc. the 27th IEEE International Conference on Computer Communications

Eighth Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA,

[3] ——, “On the throughput-delay tradeoff in cellular multicast,” in Proc. International

June 2005, pp. 1401–1406.