Prompt engineering has proven to be a crucial step in leveraging pretrained large language models (LLMs) in solving various real-world tasks. Numerous solutions have been proposed that seek to automate prompt engineering by using the model itself to edit prompts. However, the majority of state-of-the-art approaches are evaluated on tasks that require minimal prompt templates and on very large and highly capable LLMs. In contrast, solving complex tasks that require detailed information to be included in the prompt increases the amount of text that needs to be optimised. Furthermore, smaller models have been shown to be more sensitive to prompt design. To address these challenges, we propose an evolutionary search approach to automated discrete prompt optimisation consisting of two phases. In the first phase, grammar-guided genetic programming is invoked to synthesise prompt-creating programmes by searching the space of programmes populated by function compositions of syntactic, dictionary-based and LLM-based prompt-editing functions. In the second phase, local search is applied to explore the neighbourhoods of best-performing programmes in an attempt to further fine-tune their performance. Our approach outperforms three state-of-the-art prompt optimisation approaches, PromptWizard, OPRO, and RL-Prompt, on three relatively small general-purpose LLMs in four domain-specific challenging tasks. We also illustrate several examples where these benchmark methods suffer relatively severe performance degradation, while our approach improves performance in almost all task-model combinations, only incurring minimal degradation when it does not.

Grammar-Guided Genetic Programming is already outperforming humans at creating efficient transmission schedulers for large heterogeneous communications networks. We have previously proposed a multi-level grammar approach which achieved significantly better results than the canonical Grammar-Guided Genetic Programming approach. Initially, a restricted `small' grammar is utilised in order to discover suitable structures. A full grammar is then adopted after this initial phase. Hence, evolution can focus on maximising performance, by fine-tuning the well-structured models. In this work, we propose to use a hierarchical approach by employing multiple small grammars instead of a unique small grammar at the lower level, in conjunction with the full grammar at the upper level. To use multiple small grammars while maintaining the same computational budget, we have to use either (i) reduce the number of generations, or (ii) reduce the size of the population for the evolution with each of the small grammars. In this work, we confirm that the hierarchical grammar approach using the division of number of generations strategy achieves significantly better results than the multi-level approach, but requires defining an ideal number of small grammars to achieve the best performance. We also show that the hierarchical grammar approach using the division of population size strategy achieves significantly better results than the multi-level approach. However the division of population size strategy is less sensitive to the number of small grammars.

The accurate localisation and tracking of objects is crucial in many domains. In this paper, we focus on location tracking in wireless networks. Reliable localisation will be essential for self-driving, future factories, and beamforming in 5G deployments. Time-of-arrival (TOA) based localisation systems use synchronised nodes to receive radio signals sent by transmitters at the object to be located. The time it takes for the signal to propagate in a straight line to the receiver is used to trilaterate the position of the transmitter. However, the correct TOA can be difficult to identify due to multiple reflected copies of the same signal arriving at different times. This paper presents a bidirectional recurrent neural network (BiRNN) for estimating the TOA from the channel impulse response (CIR) of the signal, and a multilayer perceptron (MLP) to trilaterate the position of the transmitter from the TOA. The BiRNN and MLP are trained using measured CIR data, and outperform the conventional approaches for TOA estimation and trilateration.

Wireless communications networks are operating at breaking point during an era of relentless traffic growth. Network operators must utilize scarce and expensive wireless spectrum efficiently in order to satisfy demand. Spectrum on the links between cells and user equipments ('users': smartphones, tablets, etc.) frequently becomes congested. Capacity can be increased by transmitting data packets via multiple links. Packets can be routed through multiple Long Term Evolution (LTE) links in existing fourth generation (4G) networks. In future 5G deployments, users will be equipped to receive packets over LTE, WiFi, and millimetre wave links simultaneously.

Evolutionary computation is used to automatically evolve small cell schedulers on a realistic simulation of a 4G-LTE heterogeneous cellular network. Evolved schedulers are then further augmented by human design to improve robustness. Extensive analysis of evolved solutions and their performance across a wide range of metrics reveals evolution has uncovered a new human-competitive scheduling technique which generalises well across cells of varying sizes. Furthermore, evolved methods are shown to conform to accepted scheduling frameworks without the evolutionary process being explicitly told the form of the desired solution. Evolved solutions are shown to out-perform a human-engineered state-of-the-art benchmark by up to 50%. Finally, the approach is shown to be flexible in that tailored algorithms can be evolved for specific scenarios and corner cases, allowing network operators to create unique algorithms for different deployments, and to postpone the need for costly hardware upgrades.

Traditional single-tiered wireless communications networks cannot scale to satisfy exponentially rising demand. Operators are increasing capacity by densifying their existing macro cell deployments with co-channel small cells. However, cross-tier interference and load balancing issues present new optimization challenges in channel sharing heterogeneous networks (HetNets). One-size-fits-all heuristics for allocating resources are highly suboptimal, but designingad hoccontrollers requires significant human expertise and manual fine-tuning. In this paper, a unified, flexible, and fully automated approach for end-to-end optimization in multi-layer HetNets is presented. A hill climbing algorithm is developed for reconfiguring cells in real time in order to track dynamic traffic patterns. Schedulers for allocating spectrum to user equipment are automatically synthesized using grammar-based genetic programming. The proposed methods for configuring the HetNet and scheduling in the time–frequency domain can addressad hocobjective functions. Thus, the operator can flexibly tune the tradeoff between peak rates and fairness. Far cell edge downlink rates are increased by up to 250% compared with non-adaptive baselines. Alternatively, peak rates are increased by up to 340%. The experiments illustrate the utility and future potential of natural computing techniques in software-defined wireless communications networks.

Small Cells are being deployed alongside pre-existing Macro Cells in order to satisfy demand during the current era of exponential growth in mobile traffic. Heterogeneous networks are economical because both cell tiers share the same scarce and expensive spectrum. However, customers at cell edges experience severe cross-tier interference in channel sharing Het-Nets, resulting in poor service quality. Techniques for improving fairness globally have been developed in previous works. In this paper, a novel method for service differentiation at the level of individual customers is proposed. The proposed algorithm redistributes spectrum on a millisecond timescale, so that premium customers experience minimum downlink rates exceeding a target threshold. System level simulations indicate that downlink rate targets of at least 1 [Mbps] are always satisfied under the proposed scheme. By contrast, naive scheduling achieves the 1 [Mbps] target only 83% of the time. Quality of service can be improved for premium customers without significantly impacting global fairness metrics. Flexible service differentiation will be key to effectively monetizing the next generation of 5G wireless communications networks.

Genetic Algorithms (GAs) have been shown to be a very effective optimisation tool on a wide variety of problems. However, they are not without their drawbacks. GAs require time to run, and evolve a bespoke solution to the desired problem in real time. This requirement can prove to be prohibitive in a high-frequency dynamic environment where on-line training time is limited. Neural Networks (NNs) on the other hand can be trained at length off-line, before being deployed on-line, allowing for fast generation of solutions on demand. This study presents a hybrid approach to time-frame scheduling in a high frequency domain. A GA approach is used to generate a dataset of optimised human-competitive solutions. Deep Learning is then deployed to extract the underlying model within the GA, enabling fast optimisation on unseen data. This hybrid approach allows for NNs to generate GA-quality schedules on-line, almost 100 times faster than running the GA.

Heterogeneous cellular networks are composed of macro cells (MCs) and small cells (SCs) in which all cells occupy the same bandwidth. Provision has been made under the third generation partnership project-long term evolution framework for enhanced intercell interference coordination (eICIC) between cell tiers. Expanding on previous works, this paper instruments grammatical genetic programming to evolve control heuristics for heterogeneous networks. Three aspects of the eICIC framework are addressed including setting SC powers and selection biases, MC duty cycles, and scheduling of user equipments (UEs) at SCs. The evolved heuristics yield minimum downlink rates three times higher than a baseline method, and twice that of a state-of-the-art benchmark. Furthermore, a greater number of UEs receive transmissions under the proposed scheme than in either the baseline or benchmark cases.

Network operators are struggling to cope with exponentially increasing demand. Capacity can be increased by densifying existing Macro Cell deployments with Small Cells. The resulting two-tiered architecture is known as a Heterogeneous Network or 'HetNet'. Significant inter-tier interference in channel sharing HetNets is managed by resource interleaving in the time domain. A key task in this regard is scheduling User Equipment to receive data at Small Cells. Grammar-based Genetic Programming (GBGP) is employed to evolve models that map measurement reports to schedules on a millisecond timescale. Two different fitness functions based on evaluative and instructive feedback are compared. The former expresses an industry standard utility of downlink rates. Instructive feedback is obtained by computing highly optimised schedules offline using a Genetic Algorithm, which then act as target semantics for evolving models. This paper also compares two schemes for mapping the GBGP parse trees to Boolean schedules. Simulations show that the proposed system outperforms a state of the art benchmark and is within 17% of the estimated theoretical optimum. The impressive performance of GBGP illustrates an opportunity for the further use of evolutionary techniques in software-defined wireless communications networks.

Grammatical Evolution (GE) is applied to the problem of load balancing in heterogeneous cellular network deployments (HetNets). HetNets are multi-tiered cellular networks for which load balancing is a scalable means to maximise network capacity, assuming similar traffic from all users. This paper describes a proof of concept study in which GE is used in a genetic algorithm-like way to evolve constants which represent cell power and selection bias in order to achieve load balancing in HetNets. A fitness metric is derived to achieve load balancing both locally in sectors and globally across tiers. Initial results show promise for GE as a heuristic for load balancing. This finding motivates a more sophisticated grammar to bring enhanced Inter-Cell Interference Coordination optimisation into an evolutionary framework.