Enhanced coding of high-frequency tonal components in MPEG-D USAC through joint application of ESBR and sinusoidal modeling