Common Radio Resource Management Policy for Multimedia Traffic in Beyond 3G Heterogeneous Wireless Systems

Abstract: Beyond 3G wireless systems will be composed of a variety of Radio Access Technologies (RATs) with different, but also complementary, performance and technical characteristics. To exploit such diversity while guaranteeing the interoperability and efficient management of the different RATs, common radio resource management (CRRM) techniques need to be defined. This work proposes and evaluates a CRRM policy that simultaneously assigns to each user an adequate combination of RAT and number of radio resources within such RAT to guarantee its QoS requirements. The proposed CRRM technique is based on linear objective functions and programming tools.

• The paper introduces an integrated CRRM policy using linear programming to jointly determine optimal RAT selection and resource allocation for multimedia traffic.
• It demonstrates robust performance through Matlab simulations, ensuring high minimum and mean QoS even under increased load conditions.
• The policy employs fairness and service prioritization, guaranteeing minimum QoS for diverse services while effectively managing resource constraints.

Common Radio Resource Management for Multimedia Traffic in Beyond 3G Heterogeneous Wireless Systems

Introduction and Motivation

This paper addresses the challenge of efficient and interoperable radio resource management in Beyond 3G (B3G) heterogeneous wireless systems, composed of multiple coexisting Radio Access Technologies (RATs) with diverse technical characteristics. The authors identify the necessity for Common Radio Resource Management (CRRM) policies that jointly optimize both RAT selection and intra-RAT resource allocation. This joint approach is essential for satisfying user and operator objectives—chiefly, Quality of Service (QoS) guarantees for multimedia traffic while maximizing system efficiency. Previous studies have treated RAT selection and intra-RAT allocation as separate problems, and early CRRM proposals either neglect resource granularity for individual users or are limited to specific performance criteria. The novel contribution in this work is an integrated CRRM policy using linear programming that simultaneously determines—for each user—the optimal combination of RAT and the number of radio resources within that RAT to meet service-specific QoS requirements.

Utility-Driven Resource Assignment Model

The resource assignment relies on service-dependent utility functions, where utility captures user-perceived QoS for various resource allocations within each RAT. The system targets three representative multimedia traffic classes: email, web, and real-time H.263 video (across varying bit rates). The utility functions for these services are defined such that a utility value of zero is associated with any allocation failing to meet the service’s minimum acceptable QoS. Notably, for video services, utility corresponds to the percentage of correctly transmitted video frames—a model that is independent of the video bit rate and adaptable to transmission conditions.

The mapping from resources to utility is RAT-specific, accounting for the differences in achievable transmission rates in GPRS, EDGE, and HSDPA, and incorporates Adaptive Modulation and Coding (AMC). For each RAT, representative average data rates per timeslot/code are selected to ensure a robust tradeoff between throughput and error-correction properties. This model facilitates flexible and dynamic radio resource allocation that is sensitive to both QoS objectives and RAT capabilities.

CRRM Formulation and Optimization

The CRRM policy is posed as a Binary Integer Programming (BIP) problem. The objective is to maximize aggregated user utility, operationalized through a fairness criterion designed to equitably balance user satisfaction across heterogeneous traffic types. The precise objective function can be formally written as:

$\max \sum_{j} \ln(u_j)$

where $u_j$ is the utility achieved by user $j$ under a given RAT/resource assignment. The binary decision variable $y_{j,s}^r$ indicates whether user $j$ is assigned $s$ radio resources from RAT $r$. The formulation incorporates several service constraints: (i) resource exclusivity (a user is assigned resources from a single RAT per distribution round), (ii) resource cap (system-wide resource limits), and (iii) service prioritization (real-time video $>$ web $>$ email), which is enforced under conditions of scarcity.

The problem leverages the Branch and Bound method in conjunction with the simplex algorithm to efficiently explore the combinatorial solution space, balancing fairness and service-based prioritization.

Numerical Results and Key Insights

Simulation studies in Matlab evaluate the CRRM framework in various load scenarios drawn from two multimedia traffic profiles. Under moderate system load (e.g., 20 users per cell), the policy achieves high levels of minimum and mean QoS attainment across all service classes. For 64 kbps video, 76% of transmissions receive allocations guaranteeing the maximum attainable utility (i.e., utility = 1 corresponding to 2 HSDPA codes). As load increases to 30 users per cell or as video bitrates (and thus minimum QoS requirements) rise, the resource allocation adaptively shifts: the proportion of users satisfying only minimum QoS requirements increases, and some lower-priority services (especially email) experience a rise in service denial (no resource assigned), reflecting both resource shortages and strict prioritization.

These outcomes validate two principal claims:

• The joint CRRM approach is able to guarantee minimum QoS for the largest feasible subset of users under resource constraints, rather than maximizing the QoS for a select few.
• The policy adapts dynamically to both traffic mix and load, redistributing resources as needed to respect service priorities and fairness.

Notably, the authors do not permit simultaneous assignment of resources from multiple RATs to a single user, a design decision highlighted as a limitation and an avenue for future work.

Implications and Future Prospects

This work demonstrates the viability of linear programming-based CRRM for heterogeneous B3G systems, providing a structured framework for unified and adaptive resource assignment. Practically, such techniques can underpin operator strategies for service differentiation and efficient network utilization in environments characterized by heterogeneity both in RAT technologies and traffic types. The explicit prioritization hierarchy aligns well with current and emerging service-level agreements (SLAs) in commercial networks.

Theoretically, the provision of fairness while honoring service priorities sets a foundation for more advanced multi-objective optimization in radio resource management. Potential future directions include:

• Extension to dynamic or distributed CRRM architectures that allow inter-RAT simultaneous allocation per user.
• Incorporation of user mobility and time-varying channel statistics for enhanced adaptability.
• Integration with future wireless standards (e.g., 5G NR, 6G) and network slicing paradigms for further exploitation of RAT diversity.

Conclusion

The paper presents a formal and systematic CRRM policy for multimedia traffic in B3G heterogeneous wireless systems, leveraging linear programming for jointly optimal RAT selection and resource allocation. The policy maximizes user utility equitably across service types, while adhering to practical resource and prioritization constraints. The results demonstrate robust performance, particularly under load escalation and increasingly stringent QoS demands, underscoring the approach’s relevance for modern and future heterogeneous wireless networks.