In Blind Source Separation (BSS) the challenge is to recover the source signals from the observed mixed signals. Blindness means that neither the sources nor the mixing system are known. Separation can be based on the theoretically limiting but practically feasible assumption that the sources are statistically independent. The statistical independence of source signals assumption connects BSS and Independent Component Analysis (ICA). The main aim of this research is to solve the separation problem for source signals and mixing system that are not known by comparing two activation functions. The research uses the Natural Gradient Algorithm (NGA) to separate pairs of sub-Gaussian (music), super-Gaussian (speech) and sub-superGaussian mixed signals into their original components using Independent Component Analysis (ICA) assumption of statistical independence of the source signals. Two activation functions are used within the NGA for each of the pairs before separation comparison is made. The NGA is formulated using instantaneous Blind Signal Processing where time delay is not factored in the computation of the independent signals. The design uses a 2 x 2 Multiple Input Multiple Output (MIMO) system to accept the pairs of blind audio signals, mix them and separate them to retain their original form or their filtered version. The Fibonacci activation function and the Sigmoid activation functions are used in iterating the coefficients of the NGA up to a hundred iterations where convergence is realized. Comparing the output (estimated) to the input signals is by waveforms, frequency spectra, and the measure of the Magnitude-Squared Coherence. The results show that the NGA algorithm with Fibonacci and Sigmoid

Cassava (Manihot escutenta, Crantz), is one of the high yielding, disease and drought resistant crop that can be used as an alternative to maize in Kenya. The leaves are highly nutritious and serve as an alternative to green leafy vegetables. However, cassava leaves production, consumption and processing has been low due to the lack of a well-structured cassava value chain and standard postharvest handling practices. Additionally, the high content of anti-nutrients discourages consumption at household. This study sought to evaluate the harvesting and postharvest practices, microbial and chemical profile of fermented cassava leaves from selected Kenyan coastal varieties. A total of 247 respondents were nominated from the two counties 120 and 127 respondents in Kilifi and Taita Taveta respectively. A completely randomized experimental design was used for chemical and microbial analysis. Almost all respondents (99.6%) grew cassava for food and Kibanda meno was the most preferred variety. In both counties, farmers harvested few leaves or piecemeal by handpicking and most commonly in the morning hours. Additionally, cassava value addition was limited to drying (82.6%) and fermentation (4.1%). The respondents (65.2%) preserved cassava for a maximum of 15 days. Fermentation followed by oven-drying and sun-drying significantly (p<0.001) reduced the tannins, oxalates and cyanide to recommended levels. The sensory scores of the fermented leaves averaged at 5 points on a sevenpoint hedonic scale stating, that they were likeable in comparison to the non-fermented samples. The results also indicated that lactic acid bacteria (LAB) were the predominant microorganisms in cassava leaves fermentation. The mean log CFU of yeasts and molds, LAB and coliforms were 6.96, 7.99 and 8.70 respectively. Leaf position and cassava leaves variety significantly (p<0.001) influenced microbial load during fermentation. Since LAB is the predominant microorganism in cassava leaves fermentation there is need for isolation of its pure cultures. This study concludes that fermentation reduces the anti-nutrient content in cassava leaves making it safe for consumption thus should be adopted for value addition of cassava leaves.

ABSTRACT

Abstract

abstract

abstract

In Eastern and Southern African region maize grain prices is recorded highest in Kenya. There has been a reduction of prices of agricultural commodity from their recorded peak in 2008, but, this has not been the case in Kenyan maize prices. This then necessitates a study on transmission of world prices into Kenyan maize prices. By broadly examining the level of price transmission between the global and domestic maize markets in Kenya. Thus, this study assessed the integration of maize markets in Kenya, and evaluated the international maize prices transmission into domestic markets in Kenya. The deflated wholesale monthly price data spanning January 2002 to December 2020 was used. The data for maize prices data for Nairobi, Nakuru, Eldoret, Kisumu and Mombasa was sourced from Ministry of Agriculture, Livestock, Fisheries and Cooperatives Kenya, while the international wholesale maize prices was obtained from the FAO, GIEWS database. Co-integration, Granger Causality and Error Correction models were employed to test for market integration and price transmission. A strong integration was found between world and domestic maize markets as indicated by a high long-run elasticity of 0.60. The speed of price adjustment is relatively fast implying that 60 percent of the maize prices variations of the world are transmitted to Kenyan markets in long-run, on average it takes a month to correct 13 percent of the deviations from equilibrium. Hence, the improvement of rural areas infrastructure to accelerate movement of goods and services should be done by government of Kenya. The government should not engage in Maize trade as this distorts competitive functioning of maize markets. These results provide evidence on the need for policy makers to formulate policies which avoids market exploitations and reduce distortions. This will help in enhancing efficiency of the maize marketing system.

Maize (Zea mays L.) is a vital grain that is rich in starch and is considered the staple food of Kenya. The crop is highly productive in ecological zones that have its favorable climate; however, the grain is highly susceptible to mycotoxin contamination which is greatly attributable to poor postharvest handling practices. Little information exists on maize postharvest management especially among the small-scale farmers who are the majority in Kenya. There is lack of standard postharvest, handling, storage procedures for maize grains which contributes to huge postharvest losses due to mycotoxin contamination. The study is aimed at understanding the post-harvest knowledge of the small-scale farmers and the practices employed in their day to day activities and their impact on the aflatoxin levels on the final product. A baseline survey was conducted using semi-structured questionnaire. A total of 200 respondents were interviewed. The current findings indicate that maize farming in Trans Nzoia consisted of male (67%) as compared to female (33%) responents. Male respondents had a higher knowledge score on postharvest practices in comparison to their females counterparts, however, this was not significant (p >0.05). The level of education of respondents was significantly associated (•=37.49a , p<0.05) with their knowledge on mycotoxin contamination. More than eight in every ten respondents (83.2%) had knowledge that inadequate drying (high moisture content) of maize and long periods exposes maize to the risk of mycotoxin contamination. Majority of the respondents (49.5%) relied on the casual laborer’s as source of human labor during harvesting. Additionally, majority (88.4%) had knowledge that poor ventilation in the storage facility causes fungal contamination, mould growth and the eventual mycotoxin contamination. Moreover, most of the respondents (83%) associated mycotoxins with cause of diseases in humans. Total aflatoxin levels in the initial sample extended from 0.00 to 9.12 µg/kg, with a mean of 1.96 µg/kg. Aflatoxin levels were found to be between 3.69 and 15.43 µg/kg after two months of storage, with a mean of 2.96 µg/kg. Initial total fumonisin concentrations ranged between 0.00 and 1.36 µg/kg, with a mean of 0.44 µg/kg. After two months of storage, fumonisin levels ranged from 0.00 to 1.51 µg /kg, with a mean of 0.60 µg/kg. Only 9.68 percent and 38.71 percent of the samples, respectively, had levels of Aflatoxin and fumonisin above the WHO-recommended levels of 10ppb and 500ug/kg, respectively. The study concludes that mycotoxin contamination of maize pre and postharvest among small-scale farmers was evident in Trans Nzoia and measures should be implimented to reduce the levels of contamination. Trainings of farmers and dispatch of additional extension officers can help improve the knowledge levels of farmers.

Eucalypts account for about 39600 ha of land in Kenya while other exotic species, Cupressus and Pinus, at 28,900 and 6,800 ha of land, respectively in the year 1999 though in 2009, the total area under eucalypts cultivation stood at 100, 000 ha. Nonetheless, in 2012, wood supply was estimated to be 31 million cubic meters while the demand was about 42 million cubic meters but, future forecasting shows that wood supply is set to increase to about 36 million cubic meters and a demand of about 51 million cubic meters by the year 2032. This demand could be addressed using planation forestry with trees such as eucalypts, but there is no active breeding program in Kenya due to insufficient information on genetic diversity and pollination techniques. The objectives of this study were to: i) determine the E. grandis genetic structure from selected candidate plus trees and their progenies using microsatellite markers from five main seed collection centers; and ii) generating F1 progenies from E. grandis and E. urophylla genotypes in Muguga, Kenya as a strategy of increasing genetic diversity. The genetic fingerprinting work, entailed genomic DNA extraction using CTAB method, DNA quality check and quantification using Nano drop nucleic acid analyzer, model Shimadzu Bio Spec-nano 206-26300-48, SSR screening and reconstitution using fluorescently labeled EMBRA primers, PCR amplification using Veriti™ 96-well thermal cycler and capillary electrophoresis by Applied Biosystems 3500 Genetic Analyzer. Data on allele frequency, allelic richness, gene diversity, heterozygosity and Polymorphic Information Content (PIC) then analyzed by Power Marker v3.2.5 Genetic distance matrices were generated by GenAlEx v6.5 and used for computing Principal Coordinates Analysis (PCoA) to visualize variations and similarities. Neighbor-joining trees were constructed using DARwin v6.0.21 where the resulting trees were visualized. Variation partitioning within and among group components was computed using Analysis of Molecular Variance (AMOVA) in GenAlEx v6.5 which facilitated estimation of broad sense heritability (H2 ), standardized allelic patterns across different families and populations and components of variance significance levels using 999 permutations. From analysis, results generally indicated high genetic diversity of 0.844 suggesting high genetic variability for possible exploitation in future breeding programs. The work on pollination involved collecting ripe flowers at anthesis stage of development, pollen extraction and examination of its characteristics, in-vitro pollen germination on liquid media to determine viability, pollen shortterm storage at 4⁰C, controlled pollination by conventional and one-stop techniques and lastly seed extraction at maturity. Data was also collected on various parameters including morphological traits of pollen and mother parents, pollen germination rate on nutrient agar medium, flower diameter and length of artificially and naturally pollinated flowers. From this analysis, a success rate of around 28.6% was realized indicating that more Eucalyptus breeding programs can be established using Kenyan germplasm. Conclusively, the study confirmed that there is need for maximum utilization of genotypes from the main seed collection zones for heterosis or targeting of specific traits, and also ratified that different breeding objectives such as fast growth and disease resistance can be achieved through controlled pollination.

The study being descriptive in nature used primary and secondary data from the following sources that is questionnaire and published financial statements of the company. Prior to secondary data analysis, primary data was analysed using descriptive statistics to establish the impacts of internal controls on the company’s operational efficiencies and financial performance. Internal Control System is a vital in every organization to achieve their management objectives. The study focuses in AFFS’s. The study examined whether there is a relationship between the internal controls system and the financial performance. In this study internal control is measured by control environment, risk assessment, control activities information and communication and monitoring and the financial performance are measured by profit, efficiency and liquidity respectively. To test this hypothesis data were collected through questionnaire from the employees of AFFS’s. Regression analysis was used to measure the variables. The study finds that internal control and financial performance are statistically significant in determining financial performance. The study concluded that, the AFFS’s external auditors report their findings directly to the board or to a board committee while the board reviews the qualifications and the independence of the company's external auditors. Policies and procedures are periodically reviewed by the board to ensure that appropriate internal controls have been established. In addition, all company personnel in the areas of control understand their roles, accountability for their activities, approves the company's systems of internal controls. The study recommends that, for any company, the external auditors report their findings directly to the board or to a board committee; the board should review the qualifications and the independence of the company's external auditors. Policies and procedures should be periodically reviewed by the board to ensure that appropriate internal controls have been established.