High speed downlink packet access and beyond

By Dr. Ariela Zeira, InterDigital Communications Corporation

Due to the table in this article, we have made a PDF available for better viewing.

Third generation wireless systems were introduced to extend the data capabilities by providing quality of service (QOS) management and enabling the high data rate required for high speed data. To satisfy predicted future increasing demands on even higher data services additional enhancements are being incorporated into the different 2.5G and 3G standards including EDGE, WCDMA (FDD and TDD), and CDMA 2000 (1xEV-DV and 1xEV-DO).

This article reviews the new features recently introduced or being considered for evolving air interface standards and discusses their impact on system performance. Enhancing the performance of communication from the base station to the mobile terminal (downlink or forward link) is discussed and a short review of enhancements of transmission from the mobile terminal to the base station (uplink or reverse link) is provided.

Downlink Enhancements

Due to the asymmetric nature if data services most of the effort to date has been focused on the downlink direction. Downlink enhancements that are being incorporated or considered for evolving air interface standards are:

• Adaptive modulation and coding

• Hybrid ARQ

• Fast cell selection

• Multiple input multiple output (MIMO) antenna array processing

• Fast scheduling

Table 1 provides a summary of the implementation status of the above enhancements in wireless standards. One can immediately observe that all wireless standards evolve in a similar direction. A description of the downlink enhancement follows.

Adaptive modulation and coding - Higher order modulation (16 QAM, 64 QAM) provides higher spectral efficiency in terms of bits/sec/Hz compared to QPSK. Therefore it can be used to increase the peak data rate for a given bandwidth. For example it can enable peak data rates in the order of 10 Mbits/sec within the current 5 MHz WCDMA bandwidth.

However, higher order modulation schemes are significantly less robust to noise, interference and other channel impairments. Hence higher order modulation must be combined with fast link adaptation where the modulation scheme, coding rate and other parameters are adapted to the instantaneous channel conditions. When fast link adaptation is used, users experiencing favorable channel conditions, e.g., close to the cell site, can be assigned higher order modulation and high code rate thus achieving higher peak rates. On the other hand, users with less favorable conditions, e.g., users close to the cell border or users at a deep fade need to use the more robust QPSK modulation and low coding rates. By adapting the link parameters to the instantaneous channel conditions the downlink user throughput and cell throughput are maximized.

Hybrid ARQ - In packet data services the receiver typically detects and requests a retransmission of erroneously received packets. Until the introduction of Hybrid ARQ (also referred to as physical layer ARQ) the signaling reporting successful or unsuccessful transmission was done via higher layer signaling resulting in long delays in the retransmission process. When fast link adaptation is used — a faster ARQ mechanism is needed to add robustness to the link adaptation. Hybrid ARQ can be regarded as an implicit link adaptation technique that is not based on channel quality measurements. It is tightly coupled with and complements fast link adaptation. While fast link adaptation provides an initial estimate for the redundancy required for a reliable transmission, hybrid ARQ enables fine-tuning of the effective coding rate.

Physical layer acknowledgements are used for re-transmission decisions. Hybrid ARQ also enables additional gains by soft combing of packets from the original and subsequent packets prior to the decoding attempt. The simplest form of hybrid ARQ is Chase combining [1] where number of repeats of each coded data packet is sent. The decoder combines multiple received copies of the coded packet weighted by the SNR prior to decoding. This method provides time diversity gain and is very simple to implement. An alternative scheme of hybrid ARQ is Incremental Redundancy [2], where instead of sending simple repeats of the entire coded packet, additional redundant information is incrementally transmitted if the decoding fails on the previous attempt.

Fast cell selection - When fast cell selection is employed, the terminal makes recommendations on the "best" cell for downlink packet transmission and signals this to the network. Determination of the best cell may not only be based on radio propagation conditions but also available resources such as power and code space for the cells in the active set. Different wireless systems vary from each other on the actual implementation of fast cell selection, and in particular on whether the terminal or the network makes the decision about the cell for the next downlink packet transmission. For example, in CDMA2000 1x EV-DO the network follows the terminal's recommendation, while in CDMA2000 1x EV-DV the network makes the decisions taking in account also available resources.

Multiple Input Multiple Output (MIMO) antenna processing - MIMO techniques can be classified as evolutionary versus revolutionary techniques. Evolutionary techniques apply standard transmit and receive multiple antenna techniques (e.g. base station transmit diversity and handset receive diversity) and do not require joint transmitter-receiver design. On the other hand, revolutionary techniques are characterized by the need in joint transmitter-receiver design and require significant changes in the air interface standards. The class of revolutionary techniques includes the following:

• Joint Transmit Receive Spatio-Temporal Processing [3]

• Space-time coding [4]

• Code Reuse architecture [5]

Currently there is an on going debate in standard bodies on the relative merits of each approach (evolutionary vs. revolutionary). This debate will be resolved by simulation studies with common simulation assumptions. The main motivation for revolutionary techniques is a set of relatively recent information theoretic results indicating that the potential gains of optimal MIMO techniques substantially exceed gains achievable by evolutionary techniques [6]. It is still an open question whether these potential gains can be indeed realized in practical wireless environments.

Fast scheduling - Fast scheduling is the mechanism determining which user(s) get the cell resources at any given time interval. In some of the standards the scheduler also decides the data rates (and transmit configurations) of scheduled users. The scheduler is a key element in the design of a packet data system as it determines the overall behavior of the system. Before the introduction of fast scheduling the scheduler has been typically in the Radio Network Controllers (RNC) or the Base Station Controller (BSC). To enable fast scheduling the fast scheduler resides in the base station.

System performance studies often assume the Max C/I scheduler and the round robin scheduler. The Max C/I scheduler maximizes the cell throughput by allocating all of the cell resources to the user with the best channel conditions. This is generally considered to be an unfair scheduler, as users with unfavorable channel conditions will never be served. On the other hand, the round robin scheduler is a fair scheduler but the resulting cell throughput is significantly lower compared to the Max C/I scheduler. Hence, both the Max C/I and round robin schedulers are considered impractical. Practical schedulers must take in account the channel quality reported by the terminals as well as the time duration since the user has been served. The Proportional Fairness scheduler [7] is an example of such a practical scheduler.

Uplink Enhancements

Uplink enhancements are now being considered for the next step in the evolution of wireless systems. The motivation for uplink enhancements is the increasing importance of IP based services resulting in an increasing demand to improve capacity, coverage and delay in the uplink. Applications that can benefit from uplink enhancements include video clips, email, gaming, and video streaming.

Enhancements considered for WCDMA (Release 6 Study Items) include adaptive modulation and coding, Hybrid ARQ, Node B controlled scheduling, fast set up of dedicated channels, and shorter frame size and Improved QoS.


This article has discussed the motivation for enhancing the performance of wireless systems. It has provided a detailed review of downlink enhancements and touched on uplink enhancements. By comparing the evolution of different air interface standards one can observe that all wireless standards evolve in a similar direction and similar sets of enhancements are adopted or considered for all standards. These enhancements will extend the data capabilities of wireless systems by enabling the high data rates required for high-speed data     n.

Dr. Zeira is the Senior Director for Core Technology Development for InterDigital Communications Corporation. For more information please visit


[1] Chase, D., "Code combining. A maximum-likelihood decoding approach for combining an arbitrary number of noisy packets," IEEE trans. Commun., vol. COM-33, pp 385-393, May 1985. [2] Mandelbaum, D. M., "Adaptive-feedback coding scheme for hybrid ARQ systems," IEEE trans. Inform. Theory, vol. IT-20, pp. 388-389, May 1974. [3] Yang, J. and Roy, S., "On joint transmitter and receiver optimization for multiple-input-multiple-output (MIMO) transmission systems," IEEE trans. Comm., vol. 42, No.12, pp 3221-3231, December 1994. [4] Tarokh, V., Seshadri, N., and Calderbank, A. R., "Space-time codes fir high data rate wireless communication. Performance analysis and code construction," IEEE trans. Inform. Theory, vol. 44, pp. 744-765, March 1998. [5] Foschini, G. J., "Layered space-time architecture for wireless communication in a fading environment when using multiple antennas," Bell Lab. Tech J., 1996, 1 (2), pp. 41-59. [6] Foschini, G. J. and Gans, M. J., "On limits of wireless communications in a fading environment when using multiple antennas," Wireless Personal commun., 1998, 6, (3), pp 311-335. [7] Rentel, C. H. et al, "Comparative forward link traffic channel performance evaluation of HDR and 1XTREME systems," VTC 2002, pp. 160-164.